Requirements for leukocyte transendothelial migration via the

Requirements for leukocyte transendothelial

migration via the transmembrane chemokines

CX3CL1 and CXCL16

Von der Fakultät für Mathematik, Informatik und Naturwissenschaften der RWTH

Aachen University zur Erlangung des akademischen Grades eines Doktors der

Naturwissenschaften genehmigte Dissertation

vorgelegt von

M. Eng.

Nicole Schwarz

aus Strausberg

Berichter: Herr Privatdozent

Dr. rer. nat. Andreas Ludwig

Herr Universitätsprofessor

Dipl. Ing. Dr. Werner Baumgartner

Tag der mündlichen Prüfung: 16. März 2010

Diese Dissertation ist auf den Internetseiten der Hochschulbibliothek onlineverfügbar.

Meinen Eltern für die Unterstützung.

Es ist besser, ein paar Fragen zu stellen, als alle Antworten schon zu kennen.

James Thurber

Table of contents

1. Abstract 5......................................................................................................................

2. Introduction 7..............................................................................................................

2.1 Current model of leukocyte extravasation 7............................................................

2.2 Chemokines 10..........................................................................................................

2.3 Chemokine receptors 12............................................................................................

2.4 CX3CL1 and its receptor CX3CR1 14......................................................................

2.5 CXCL16 and its receptor CXCR6 21........................................................................

2.6 ADAMs 23................................................................................................................

3. Aim of this thesis 25......................................................................................................

4. Materials and Methods 26............................................................................................

4.1 Materials 26...............................................................................................................

4.2 Methods 31................................................................................................................

5. Results 48.......................................................................................................................

5.1 CX3CR1-CX3CL1 interaction 48.............................................................................

5.2 Model system for CX3CR1-CX3CL1 function 55....................................................

5.3 Molecular analysis of CX3CR1-CX3CL1 function 62.............................................

5.4 Involvement of ADAMs in CX3CR1-CX3CL1 function 74.....................................

5.5 CXCR6-CXCL16 function 80...................................................................................

6. Discussion 88.................................................................................................................

7. Literature 103...................................................................................................................

8. List of figures 115.............................................................................................................

9. List of tables 117..............................................................................................................

10. Abbreviations 118.........................................................................................................

11. Vectors 123....................................................................................................................

11.1 hCX3CR1 in pcDNA3.1+ 123....................................................................................

11.2 hCX3CR1 R127N in pcDNA3.1+ 124.......................................................................

11.3 hCX3CR1 N289A in pcDNA3.1+ 125.......................................................................

Table of contents

3

11.3 hCX3CR1 Y293A in pcDNA3.1+ 126.......................................................................

11.4 hCX3CR1 S319X in pcDNA3.1+ 127........................................................................

11.5 hCXCR6 in pcDNA3.1+ 128......................................................................................

11.6 hCXCR6 R127N in pcDNA3.1+ 129.........................................................................

11.7 hCXCR6 F128Y in pcDNA3.1+ 130..........................................................................

12. Curriculum Vitae 131...................................................................................................

13. Publications 132............................................................................................................

14. Declaration 134.............................................................................................................

15. Acknowledgements 135................................................................................................

Table of contents

4

1 Abstract

The chemokines CX3CL1 and CXCL16 and their receptors CX3CR1 and CXCR6 are

described in vascular inflammation and inflammatory cell recruitment. CX3CL1 and

CXCL16 are transmembrane surface proteins on endothelial cells inducing firm adhesion

of leukocytes via the interaction with their receptors. After shedding from the cell surface

by the metalloproteinases ADAM10 and ADAM17, they act as soluble chemoattractants

for CX3CR1- and CXCR6-expressing leukocytes, respectively.

Here, it was demonstrated that expression of transmembrane CX3CL1 on endothelial

cells promotes leukocyte transendothelial migration, and details of the underlying

mechanisms using mutated CX3CR1 variants were elucidated. The DRY motif required

for Gi-protein-coupling was mutated to DNY, which abolished the intracellular calcium

release in response to CX3CL1, but did neither affect CX3CL1 binding nor uptake.

Truncation of the C-terminus reduced ligand uptake, but not ligand binding and calcium

responses. Both variants effectively mediated firm cell adhesion, but not chemotaxis

towards soluble CX3CL1. Furthermore, they failed to induce transmigration, but

mediated retention of leukocytes on the CX3CL1-expressing cell layer. Pharmacologic

and transcriptional inhibition of ADAM10 led to reduced shedding of transmembrane

CX3CL1, which was associated with an almost complete suppression of transmigration

in response to transmembrane CX3CL1. These results indicate a multistep process of

leukocyte recruitment by transmembrane CX3CL1 involving adhesion, signaling,

initiation of transmigration, and finally proteolytic release of the transmigrating

leukocytes.

In contrast, transmembrane CXCL16 did not promote adhesion of CXCR6-expressing

cells, while soluble CXCL16 mediated chemotaxis. CXCR6 bears a DRF instead of the

DRY motif. Since this mutation is implicated in the constitutive activity of other

receptors, the DRF of CXCR6 was changed into DRY and DNF. Reconstitution of the

DRY motif did not affect ligand binding and resulted in a slight decrease in calcium

1 Abstract

5

signaling, whereas the mutation into DNF abolished calcium signaling. Both mutated

receptors still failed to induce adhesion to CXCL16-expressing cells. Signaling seemed

to depend on the arginine residue but not on the tyrosine/phenylalanine residue in DRY/

F. These results indicate that CXCL16 predominantly functions as a soluble chemokine.

Furthermore, cell recruitment by transmembrane chemokines differs. While CX3CL1

induces signaling-independent adhesion and signaling-dependent transmigration,

CXCL16 does not induce adhesion, but chemotaxis.

1 Abstract

6

2 Introduction

Inflammation is a fundamental defense reaction caused by tissue damage or injury. Its

primary purpose is the protection of the organism by removing or neutralizing injurious

agents and repairing the surrounding tissue. The inflammatory response involves three

major stages: first, dilation of the blood vessel leading to an increase of blood flow;

second, structural changes in the microvascular system; and third, localized recruitment

of various leukocyte subsets to sites of inflammation. The current model of leukocyte

extravasation from the vasculature into inflamed tissue comprises several steps governed

by diverse molecules such as cytokines, adhesion molecules and chemokines acting as

either soluble mediators, membrane-expressed ligands or signal transducing receptors.

2.1 Current model of leukocyte extravasation

Leukocyte extravasation is a multistep process that consists of at least five major

stages, and is halted when any one of them is suppressed (see Figure 1). On recognition

of pathogens, resident cells like macrophages or dendritic cells undergo activation and

release pro-inflammatory cytokines like IL-1, TNFα and chemokines. Endothelial cells

of blood vessels near the site of infection start to express cellular adhesion molecules,

such as selectins, as a result of activation by these cytokines.

The selectins (P, E, and L) are type 1 transmembrane glycoproteins that bind to

modified sialyl Lewis X (sLex) present in their ligands in a Ca2+-dependent fashion. L-

selectin is expressed by most leukocytes, whereas the E and P forms are expressed on

endothelial cells that were activated by proinflammatory stimuli. Binding of endothelial

E- and P-selectins to their corresponding ligands on the leukocytes slows down the

velocity of leukocytes in the bloodstream, leading to a rolling movement of the cells on

the vascular wall (reviewed in: [Barreiro et al., 2004]).

As leukocytes start tethering to the vascular endothelium and their rolling velocity

2 Introduction

7

slows, integrins are activated upon encountering immobilized chemokines and integrin

ligands exposed on the apical endothelial surface [Campbell et al., 1998]. This activation

step enables the arrest of leukocytes and their subsequent firm adhesion to the

endothelium under physiological flow conditions.

Figure 1: Multiple steps are necessary for leukocyte extravasation. Leukocytesrecognize and bind selectins that are expressed by cytokine-activated endothelialcells leading to a rolling movement. Subsequently, integrins become activated andundergo a conformational change. In the next step integrins, adhesion moleculesand chemokine receptors bind to their respective ligands on the endothelial cell sur-face, leading to firm adhesion. Then, leukocytes crawl on the cell surface in searchof an appropriate site for transmigration. After protrusion between two adjacent en-dothelial cells transmigration towards an increasing chemokine-gradient occurs.Figure modified after: [Man et al., 2007].

Integrins comprise a family of 24 heterodimeric receptors, each of which is composed

of an α-subunit and a β-subunit. These molecules dynamically alter their adhesive

properties through conformational changes [Beglova et al., 2002; Nishida et al., 2006].

The most relevant integrins for leukocyte adhesion to the endothelium are members of

2 Introduction

8

the β2 subfamily, particularly LFA-1 (αLβ2), as well as VLA-4 (α4β1).

Most of the ligands are transmembrane proteins that belong to the immunoglobulin

superfamily, such as endothelial intercellular adhesion molecule-1 (ICAM-1, ligand to

LFA-1) and vascular cell adhesion molecule-1 (VCAM-1, ligand to VLA-4) [Gahmberg

et al., 1990].

Chemoattractants stimulate the directional leukocyte migration called chemotaxis.

Chemokines are small chemotactic polypeptides that induce leukocyte migration towards

increasing concentration gradients of chemokines. However, soluble chemokine

gradients are unlikely to exist at the luminal endothelial surface, since they could be

easily washed away. Therefore, it has been proposed that chemokines could act as bound

form to exert proadhesive and migratory effects on leukocytes in the lumen of blood

vessels. There are several chemokines that are known to be immobilized on the

endothelial cell surface by binding to glycosaminoglycans, like IL-8 and RANTES [Rot

et al., 1992; Tanaka et al., 1993]. Additionally, there exist two special chemokines

termed CX3CL1 and CXCL16 that are produced as membrane-anchored molecules by

the endothelium, and can be shed to a soluble form by metalloproteinases of the ADAM

(a disintegrin and metalloprotease) family [Bazan et al., 1997; Garton et al., 2001;

Hundhausen et al., 2003]. CX3CL1 and CXCL16 likely play a role as transmembrane

adhesion molecules as well as soluble chemoattractants.

Subsequent to firm adhesion, leukocytes transmigrate through the endothelium without

irreversibly impairing its integrity. Two routes exist that the cells can use: they can either

move between the endothelial cells (paracellular) or they can migrate through an

endothelial cell (transcellular). It is still unclear to which extent each of the ways is

involved in transmigration. For paracellular transmigration, leukocytes encounter tight

junctions and adherence junctions and engage other transmembrane receptors, including

junctional molecules such as platelet/endothelial cell adhesion molecule-1 (PECAM-1),

members of the junctional adhesion molecule (JAM) family, and CD99, which contribute

to sequential steps of the transmigration process [Muller et al., 1993; Martin-Padura et

2 Introduction

9

al., 1998; Schenkel et al., 2002].

Transmigration requires a morphological change of leukocytes. They become

polarized with at least two regions that can be identified: the cell front and the cellular

uropod [del Pozo et al., 1995]. This allows the cell to coordinate the intracellular forces

that are needed for directed cell crawling [Geiger and Bershadsky, 2002]. Once the

leukocyte has encountered an appropriate site for transmigration, it extends pseudopods

between two adjacent cells. During the transmigration process, chemokine receptors

polarize on the leading edge, resulting in a functional specialization of this cell domain in

signal transduction [Nieto et al., 1997]. Since transmigration is mediated by intracellular

phosphorylation signals through chemokine receptors, this re-organization may act as a

sensor mechanism for chemokine-guided cell trafficking [Bokoch, 1995].

2.2 Chemokines

The superfamily of chemokines consists of at least 42 small chemotactic cytokines.

They share structural characteristics such as the size (8-10 kDa) and the presence of

conserved cysteine residues in the N-terminal region forming their 3-dimensional shape.

Chemokines are classified into four groups according to the number and position of these

cysteine residues. CXC-chemokines are characterized by the presence of one amino acid

between the first two cysteines, whereas in CC-chemokines the first two cysteines are

adjacent to each other. The C-chemokine family only contains two of the four conserved

cysteine residues, and in the CX3C-chemokine family, with its only member CX3CL1,

three amino acids separate the first two cysteines (see Figure 2).

Chemokines consist of an elongated N-terminus that includes the first cysteine.

Following the first two cysteines, the structure forms a so-called N-loop consisting of

approximately ten residues that is succeeded by a sequence of α-helices and two

antiparallel β-sheets interrupted by loops. The so-called 30- and 50-loops contain the

third and fourth cysteine residue. Due to disulfide bonds between the first and the third

2 Introduction

10

and between the second and the fourth cysteine the molecule is stabilized. The sequence

identities between different chemokines vary from less than 20% to over 90%.

Figure 2: Chemokines are divided into four groups based on their first twocystein residues. Chemokines are subdivided into four groups based on the posi-tioning of the first two conserved cysteine residues. In CXC α chemokine family,one amino acid is located between the first two chemokines; in CC β chemokinefamily, the first two cysteines are adjacent to each other; in CX3C δ chemokinefamily, there are three interfering amino acids; in C γ chemokine family, only twoof the conserved cysteine residues are present. Figure modified after: [Rostene etal., 2007].

The three-dimensional structure of each monomer is virtually identical, but the

quaternary structure of chemokines differs for each subfamily. Structural studies reveal

2 Introduction

11

several regions of chemokines to be involved in receptor binding and function, with the

N-terminal region playing an important role [Onuffer and Horuk, 2002; Allen et al.,

2007].

Some chemokines are considered pro-inflammatory and can be induced during an

immune response to promote cells of the immune system to a site of infection, while

others are considered homeostatic and are involved in controlling the migration of cells

during normal processes of tissue maintenance or development.

2.3 Chemokine receptors

Chemokines exert their biological effects by interacting with G-protein-linked 7-

transmembrane-spanning receptors, called chemokine receptors that are selectively found

on the surfaces of their target cells. These chemokine receptors are part of a much bigger

superfamily of G-protein-coupled receptors that include receptors for hormones,

neurotransmitters, paracrine substances, inflammatory mediators, certain proteases, taste

and odorant molecules, and even photons and calcium ions.

G-protein-coupled receptors are associated with a heterotrimeric G-protein (guanine

nucleotide-binding protein) consisting of three subunits (α, β and γ). On conformational

changes by ligand-binding, the receptor can act as guanine nucleotide exchange factor

and activate the G-protein by exchanging its bound GDP to GTP. The α-subunit with the

bound GTP then dissociates from the β- and γ-subunits and further affects intracellular

signaling depending on the α-subunit type.

There are two general signal transduction pathways involved in G-protein-coupled

receptors: cAMP signal pathway and phosphatidylinositol signal pathway. Gαi, or

inhibitory regulative G-protein, associated signaling leads to an inhibition of adenylate

cyclase that catalyzes the formation of cAMP from ATP. In phosphatidylinositol

signaling, phosphatidylinositol 4,5-bisphosphate (PIP2) is cleaved into the second

messengers inositol 1,4,5-trisphosphate (IP3) and diacylglycerol (DAG) by

2 Introduction

12

phospholipase C. IP3 then activates Ca2+-release channels in endoplasmatic reticulum

membranes, where it can bind to calmodulin or can be used to activate protein kinase C

together with DAG. Protein kinase C, in turn, catalyzes the phosphorylation of several

cellular proteins, resulting in altered protein activity.

GPCRs become desensitized after prolonged exposition to their ligands. G-protein

receptor kinases can phosphorylate the intracellular part of active receptors, leading to

the binding of β-arrestin that either prevents the receptor from binding to its ligand or, as

in the most cases, promotes the removal of the receptor from the membrane via clathrin-

mediated endocytosis, called internalization. When the receptor is dephosphorylated in

the acidic microenvironment, it can be re-expressed at the cell surface [Krueger et al.,

1997; Laporte et al., 1999]

G-protein-coupled receptors are grouped into six classes according to their sequence

homology and functional similarity, with the largest group being class A or rhodopsin-

like GPCRs. Chemokine receptors belong to the group of class A GPCR. To date 19

distinct chemokine receptors have been described in mammals, and they are thought to

predominantly couple to the Gαi -subunit.

Chemokine receptors are further subdivided into different families that correspond to

the 4 distinct subfamilies of chemokines they bind (CXC chemokine receptors, CC

chemokine receptors, CX3C chemokine receptors and XC chemokine receptors). Each of

the chemokine receptors binds only a single class of chemokines, although they can bind

several members of the same class with high affinity. Furthermore, several chemokines

can bind and activate a number of chemokine receptors.

2.3.1 Structural characteristics

Chemokine receptors measure approximately 350 amino acids in length. A short

extracellular N-terminus is acidic overall and may be sulfated on tyrosine residues and

contains N-linked glycosylation sites. Seven α-helical transmembrane domains - with

2 Introduction

13

three intracellular and three extracellular connecting loops composed of hydrophilic

amino acids - are oriented perpendicularly to the plasma membrane. A disulfide bond

links highly conserved cysteines in extracellular loops 1 and 2. G-proteins are coupled

through the C-terminus segment and possibly through the third intracellular loop

[Murdoch and Finn, 2000].

Throughout the GPCR super-family, several highly conserved motifs have been

identified. The aspartate-arginine-tyrosine (DRY) sequence in the second intracellular

loop is required for activation of Gi-proteins, whereas the NPX2-3Y motif located in the

seventh transmembrane region of most GPCRs contributes to ligand binding, activation

and internalization of the receptor [Slice et al., 1994; Arora et al., 1995]. Additionally,

GPCRs typically carry several serine residues within the intracellular C-terminal region,

which can become phosphorylated by G-protein-coupled receptor kinases and mediate

interaction with β-arrestins and desensitization of the receptor towards its ligand [Kim

and Caron, 2008]. Table 1 shows the conserved motifs mentioned above and their

functional relevance as shown for other GPCRs in which the motifs were altered.

2.4 CX3CL1 and its receptor CX3CR1

2.4.1 Fractalkine

CX3CL1 (fractalkine) is expressed as a cell surface protein on dendritic, epithelial,

neuronal and most prominently on endothelial cells [Bazan et al., 1997]. CX3CL1 (like

CXCL16) is an unique member of the chemokine family as it occurs in a membrane-

tethered and a soluble form. It is expressed as type I transmembrane protein (N-terminus

extracellular), and its ectodomain can be proteolytically cleaved from the cell surface.

This process is known as shedding. ADAM10 is implicated in constitutive shedding of

fractalkine [Hundhausen et al., 2003], whereas ADAM17 mediates PMA-induced

shedding [Garton et al., 2001].

2 Introduction

14

Table 1: Conserved motifs in GPCR class A. Conserved motifs in the G-protein-coupled receptor class A were mutated in a variety of receptors. The table showsthe consequence of altering motifs.

Motif Receptor Consequence ReferenceDRY rhodopsin impaired stimulation of

transducin[Acharya and Karnik, 1996]

gonadotropin-releasing hormone receptor

impaired internalization and IP3 response

[Arora et al., 1997]

CXCR4 loss of activity [Berchiche et al., 2007]oxytocin receptor constitutive activity [Fanelli et al., 1999]

CCR5 loss of activity [Gosling et al., 1997]CCR5 loss of G-protein

activation[Lagane et al., 2005]

mCX3CR1 impaired signaling and chemotaxis

[Haskell et al., 1999]

dopamine D2 and D3 receptors

impaired signaling [Kim et al., 2008]

alpha 1B adrenergic receptor

no G-protein coupling, constitutive desensitization

[Wilbanks et al., 2002]

angiotensin II type 1A receptor

no G-protein coupling, constitutive desensitization

[Wilbanks et al., 2002]

vasopressin receptor non-signaling, constitutivedesensitization

[Barak et al., 2001]

V2 vasopressin receptor no G-protein stimulation [Rosenthal et al., 1993]M1/M2 muscarinic receptor

decrease in signaling [Zhu et al., 1994]

alpha1B-adrenergic receptor

loss of receptor-mediated response

[Scheer et al., 2000]

histamine H2 receptor loss of basal activity [Alewijnse et al., 2000]NPX2-3Y beta-1-adrenergic

receptorimpaired internalization and binding

[Barak et al., 1995]

angiotensin II receptor impaired binding and sig-nal transduction

[Hunyady et al., 1995]

gastrin-releasing hor-mone receptor

no impairments [Slice et al., 1994]

truncation IL8Rbeta migration impairment [Ben-Baruch et al., 1995]CCR5 sustained calcium

response[Kraft et al., 2001]

2 Introduction

15

The membrane-tethered form consists of a chemokine domain anchored to the plasma

membrane through an extended mucin-like stalk, a transmembrane helix and an

intracellular domain. Analysis of the C-terminal cleavage fragments, which remain in the

cell membrane, reveals multiple cleavage sites used by ADAM10 [Hundhausen et al.,

2007].

Compared to other chemokines, CX3CL1 has at least three unique structural features

that may mediate its function as an adhesion molecule. It is a transmembrane molecule

with a cytoplasmic tail that may participate in signal transduction, it has a mucin domain,

and it has a three-dimensional structure that is slightly different from other chemokines

[Mizoue et al., 1999]. The fractalkine mucin domain is a 26 nm long stalk that functions

in extending the chemokine domain away from the cell surface and is functionally

similar to the stalk regions of the selectin family of molecules. It does not contribute to

fractalkine binding affinity and efficacy for stimulating calcium mobilization in

CX3CR1-expressing cells [Harrison et al., 2001].

2.4.2 CX3CR1

CX3CL1 binds to a single receptor (CX3CR1, also known as V28) that is expressed

on several leukocyte subsets including monocytes, NK cells, T cell populations, dendritic

cells and microglia [Imai et al., 1997; Combadiere et al., 1998; Dichmann et al., 2001].

CX3CR1 belongs, as all chemokine receptors, to the group of G-protein-coupled

heptahelical receptors and couples to Gi/o [Wong, 2003].

The CX3CR1 gene is multi-exonic with the ORF residing entirely in a single exon,

like most chemokine receptors. There are three functionally independent promoter

regions that direct CX3CR1 transcription, giving rise to three different transcripts that

may be expressed selectively in different cell populations [Garin et al., 2002]. Four

naturally occurring receptor variants (T57A, V122I, V249I and T280M) are described in

humans [Faure et al., 2000]. While M280 is never found in the absence of I249, the

2 Introduction

16

converse is not true [McDermott et al., 2003]. The I249 allele has been associated with

reduced fractalkine binding affinity as well as with reduced risk for acute coronary

events and improved endothelium-dependent vasodilation [McDermott et al., 2001;

Moatti et al., 2001].

Figure 3: Schematic view of CX3CR1. CX3CR1 is a seven-transmembrane G-protein-coupled receptor belonging to the family of chemokine receptors. Targetedmutation of selected acidic amino acid residues demonstrated that the binding activ-ity of CX3CR1 is critically dependent on the two negatively charged residuesAsp25 and Glu254 located on the N-terminal domain and third extracellular loop,respectively. Furthermore, the initial interaction triggers the engagement of Glu13,Asp16, and Asp266, which are necessary for CX3CR1 activation [Chen et al.,2006]. The numbers specify the positions of the various amino acid residues in theprotein. The transmembrane domains are illustrated by the cylindrical arrangementsof the amino acids.

2 Introduction

17

2.4.3 Consequences of CX3CR1-CX3CL1 interaction

Transmembrane as well as soluble CX3CL1 binds to CX3CR1. However, both

CX3CL1 variants mediate different steps of the recruitment process and employ distinct

mechanisms. Transmembrane CX3CL1 serves as an adhesion molecule, mediating

enhanced leukocyte adhesion to endothelial cells under flow conditions. This activity is

largely independent of CX3CR1-mediated Gi-protein activation and predominantly a

result of the physical interaction of the transmembrane chemokine with its receptor

[Fong et al., 1998]. By contrast, soluble CX3CL1 acts as a chemoattractant inducing

directional cell migration via CX3CR1 signaling and activation of Gi -proteins [Imai et

al., 1997].

Fractalkine expression is upregulated after stimulation with IL-1β, LPS, IFNγ and

TNFα, and it can induce its own expression, which is both mediated via the NF-κB

pathway [Garcia et al., 2000]. Treatment with CX3CL1 increases endothelial cell

proliferation in a dose- and time-dependent manner [Lee et al., 2006], and the molecule

may serve as survival factor for microglial cells and monocytes [Boehme et al., 2000;

Landsman et al., 2009].

Receptor activation by soluble fractalkine induces activation of Akt and p53, ERK and

eNOS phosphorylation in a time- and dose-dependent manner, and nuclear translocation

of NF-κB. It does not induce SAPK/JNK and p38 phosphorylation. Furthermore,

stimulation with CX3CL1 induces a calcium flux response in CX3CR1 expressing cells

in a concentration-dependent manner. The G-protein-coupled receptor inhibitor pertussis

toxin (PTX) efficiently blocks fractalkine-induced phosphorylation of ERK, Akt and

eNOS, suggesting that the receptor is linked to Gi-proteins [Lee et al., 2006].

2 Introduction

18

Figure 4: Schematic view of CX3CR1 signaling. On binding of CX3CL1 to itsreceptor CX3CR1, G-protein subunit α uncouples, and PI3-kinase (PI3K) and Phos-pholipase C (PLC) become activated. PLC activation leads to diacylglycerol (DAG)and inositol 1,4,5-trisphosphate (IP3) formation, which stimulate calcium releasefrom the endoplasmatic reticulum and subsequent activation of calcineurin, proteinkinase C (PKC) and endothelial nitric oxide synthase (eNOS). PI3K activationleads to activation of mitogen-activated protein kinase (MAP kinase) pathway andAkt and subsequent translocation of NF-κB to the nucleus.

2.4.4 CX3CL1 and CX3CR1 in clinical disease

Accumulating evidence suggests that CX3CL1-CX3CR1 interaction contributes to the

development and the progression of many inflammatory diseases, some of which are

described in more detail in the following.

Fractalkine is suggested to be involved in atherosclerosis and cardiovascular patho-

2 Introduction

19

physiology. In some human advanced atherosclerotic lesions, high levels of fractalkine

mRNA expression have been observed [Greaves et al., 2001]. Apolipoprotein E-deficient

(apoE-/-) mice show upregulated CX3CL1 expression, whereas apoE-/- mice crossing

with CX3CR1-/- mice resulted in a decreased atherosclerotic lesion formation

[Combadière et al., 2003; Lesnik et al., 2003]. Gene polymorphisms V249I and T280M

of human CX3CR1 have been reported to be a genetic risk factor for coronary artery

disease, whereas CX3CR1-V249I/T280M heterozygosity is associated with a reduced

risk of acute coronary events [McDermott et al., 2001; Moatti et al., 2001].

Additionally, fractalkine has been reported to play a role in human renal diseases

(glomerulonephritis, renal tumours, and renal transplants) and in kidney disease in

animal models. The expression of fractalkine and the presence of CX3CR1-expressing

cells have been demonstrated in patients with various types of nephropathies [Furuichi et

al., 2001; Cockwell et al., 2002]. Anti-CX3CR1 antibody treatment blocked leukocyte

infiltration into the glomeruli and improved renal function, suggesting a role for

fractalkine and CX3CR1-expressing cells in the pathogenesis of human

glomerulonephritis [Feng et al., 1999].

Furthermore, an increased expression of fractalkine has been detected in lymph nodes

and brain tissue from patients with HIV. The increased expression of CX3CL1 is

associated with a protection of neurons from two HIV-1 neurotoxins, but induces the

depletion of CX3CR1-positive T-helper cells by contact with dendritic cells [Tong et al.,

2000; Foussat et al., 2001]. Since HIV-infected patients homozygous for CX3CR1-

I249M280 progress to AIDS more rapidly than do those with other haplotypes, it has

been concluded that the specific polymorphism, CX3CR1-I249M280, is a genetic risk

factor in HIV infection [Faure et al., 2000].

It still remains unclear whether these diseases are mediated by either the

transmembrane, the soluble or both variants of CX3CL1. The investigation of each

variant is complicated by the fact that transmembrane CX3CL1 is converted into soluble

CX3CL1 by limited proteolysis. This process is called shedding and involves several cell

2 Introduction

20

surface enzymes called sheddases. A class of sheddases implicated in the shedding

process of CX3CL1 is the ADAM (a disintegrin and metalloproteinase) family with its

members ADAM10 (kuzbanian) and ADAM17 (TACE for TNFα converting enzyme).

2.5 CXCL16 and its receptor CXCR6

2.5.1 CXCL16

CXCL16 is a CXC α chemokine, but shares characteristics of CC chemokines (strong

homology to MIP-1β, long loop between Cys2 and Cys3 like CTAK and TECK, and 6

Cys-residues like 6-C-Kine), and shows structural similarities to CX3CL1 like a

transmembrane region and a chemokine domain suspended by a mucin-like stalk

[Wilbanks et al., 2001]. Initially, CXCL16 was described as scavenger receptor for

oxidized low density lipoprotein (ox-LDL), phosphatidylserine, dextran sulphate and

bacteria under the name of SR-PSOX (scavenger receptor that binds phosphatidylserine

and oxidized lipoprotein) [Shimaoka et al., 2000]. SR-PSOX specifically binds,

internalizes and degrades ox-LDL leading to foam cell transformation implicated in the

process of atherogenesis [Ross, 1993].

CXCL16 is predominantly expressed by antigen-presenting cells, including subsets of

CD19+ B cells, CD14+ monocytes/macrophages, dendritic cells, aortic smooth muscle

cells and by cells in the splenic red pulp [Matloubian et al., 2000; Wilbanks et al., 2001;

Chandrasekar et al., 2004]. Like fractalkine, CXCL16 can be shed by ADAM10 to form

a soluble chemokine that induces chemotaxis of activated T-cells and bone marrow

plasma cells via its receptor CXCR6 [Matloubian et al., 2000; Nakayama et al., 2003;

Abel et al., 2004].

2 Introduction

21

2.5.2 CXCR6

CXCR6 (also called Bonzo), like all chemokine receptors a G-protein-coupled

heptahelical receptor, is the only known receptor for CXCL16. It is expressed on CD4+

T-helper 1, CD8+ T-cytotoxic and T-regulatory 1 subsets of T-cells, smooth muscle

cells, dendritic cells, B-cells, macrophages, and subsets of natural killer cells

[Matloubian et al., 2000; Sharron et al., 2000; Kim et al., 2001; Wilbanks et al., 2001;

Hofnagel et al., 2002; Sato et al., 2005]. CXCR6 was firstly described as a fusion co-

factor for HIV-1/SIV [Deng et al., 1997], and the polymorphism CXCR6-3E/K is

implicated in a prolonged survival time from Pneumocystis carinii pneumonia in HIV-

positive individuals. In contrast, patients with the polymorphism showed impaired viral

suppression after initiation of Highly Active Anti-Retroviral Therapy (HAART) [Passam

et al., 2007].

Instead of the conserved DRY-motif that is common for G-protein-coupled receptors,

CXCR6 has a DRF-motif. This alteration was already described for the arthropod 5-

HT2βPan receptor and was implicated in its constitutive activity [Clark et al., 2004]. The

conserved NPX2-3Y motif is unaltered.

2.5.3 Consequences of CXCL16-CXCR6 interaction

CXCL16 mediates adhesion, chemotaxis and calcium signaling in CXCR6-expressing

cells [Wilbanks et al., 2001]. As described for CX3CL1, both variants - soluble and

transmembrane - might mediate different steps in the process of cell extravasation.

CXCL16 is a potent activator of NF-κB, but fails to regulate its own expression,

indicating that CXCL16, unlike CX3CL1, is not a NF-κB-responsive chemokine. Both

IL-1 and TNFα fail to influence CXCL16 expression in human aortic smooth muscle

cells and umbilical vein endothelial cells [Hofnagel et al., 2002].

CXCR6 signals via PI3K, Akt, IκK and IκB phosphorylation. Binding of CXCL16

2 Introduction

22

mediates NF-κB activation via Akt/PKB, leading to increased proliferation of human

aortic smooth muscle cells. Therefore, treatment with CXCL16 does not induce cell

death, but promotes cell survival and proliferation. Additionally, CXCL16 binding leads

to Bad (Bcl-2-Antagonist of Cell Death)-phosphorylation, indicating its role as anti-

apoptotic chemokine. This suggests that chemokines not only play a role in extravasation

processes, but also in cell survival and angiogenesis [Chandrasekar et al., 2004].

2.6 ADAMs

ADAMs are type I transmembrane proteins and belong to the metzincin superfamily of

zinc-dependent proteases. They generally consist of about 750-800 amino acids. The N-

terminal extracellular pro-domain is followed by a highly conserved metalloprotease

domain. The active site for the proteolytic cleavage contains HEXGHNLGXXHD and

contains a catalytically essential zinc-ion, which is chelated by the three conserved

histidin-residues. A conserved methionine residue beneath the active site metal, as part of

a ”Met-turn”, loops the polypeptide chain beneath the catalytic zinc ion, forming a

hydrophobic floor to the Zn2+ ion binding site [Stocker and Bode, 1995]. The disintegrin-

domain might be a ligand for integrins and is succeeded by a cysteine-rich potential cell

fusion region. The C-terminal cytoplasmic domain does not share significant sequence

similarities. Some of them contain putative consensus sequences for binding to sarcoma

(Src) and Src-related SH3 domains [Wolfsberg and White, 1996].

Although a single sheddase may shed a variety of substances, multiple sheddases can

cleave the same substrate resulting in different consequences. More than 40 type-1 cell

surface proteins are known substrates for ADAMs (reviewed in: [Pruessmeyer and

Ludwig, 2008]). CX3CL1 can be cleaved from the cell surface by two distinct enzyme

activities. ADAM10 is implicated in the constitutive shedding of CX3CL1 of

unstimulated cells, whereas cleavage by ADAM17 is inducible by phorbol esters such as

phorbol 12-myristate 13-acetate (PMA) [Garton et al., 2001; Hundhausen et al., 2003].

2 Introduction

23

ADAM10 and ADAM17 are upregulated in activated endothelium and play a role in

ectodomain shedding of adhesion molecules during leukocyte recruitment [Boulday et

al., 2001]. Since CX3CL1 acts as an adhesion molecule that is upregulated in activated

endothelium, it can be cleaved by ADAM10 or 17 to form an additional soluble

chemoattractant to recruit more inflammatory cells as well as to allow abrogation of firm

adhesion in order to let transmigration take place.

2 Introduction

24

3 Aim of this thesis

This thesis aimed to characterize the role of CX3CL1 in distinct steps of leukocyte

recruitment and to elucidate the mechanism of CX3CL1-mediated cell recruitment.

Primary cells (PBMC) were investigated for adhesion and chemotaxis in response to

soluble and transmembrane CX3CL1 expression on endothelial cells. Then, a model

system of cell lines for adhesion and chemotaxis/transmigration induced by CX3CL1

was established and confirmed to show the same behavior as primary cells.

In order to determine which structural characteristics are important, mutated receptor

variants altering the DRY motif in the second intracellular loop, the NPX2-3Y motif in

the seventh transmembrane domain, and deleting all C-terminal serine residues by

truncation were generated. The effect of mutating these highly conserved regions was

studied in functional assays like binding, ligand uptake, signaling, adhesion,

transmigration, and pseudopod formation.

Since the activity of ADAM10 and 17 is supposed to be required in efficient leukocyte

extravasation, it was examined whether the metalloproteases ADAM10 and 17 are

required in CX3CL1-mediated function using pharmacological inhibitors to the proteases

and shRNA silencing.

CXCL16 is the only other chemokine that is expressed as transmembrane as well as

soluble form. This raised the question whether the findings concerning CX3CR1 are

similar for CXCR6. Therefore, the CXCL16-CXCR6 interaction was examined in

functional assays like adhesion and chemotaxis. Furthermore, the unusual DRF motif of

CXCR6 was altered and the functional consequence of the receptor variants was studied.

3 Aim of this thesis

25

4 Materials and Methods

4.1 Materials

4.1.1 Cell Culture

Media, supplements and reagents used for cell culture were purchased from PAA

(Pasching, Austria) if not stated otherwise. Cells were cultured on cell culture

plasticware from Sarstedt (Nuembrecht, Germany).

4.1.2 Primer

All primers were ordered from Eurofins MWG Operon (Ebersberg, Germany). They

were diluted to a final concentration of 100 pmol/µl.

Primer Sequence

hCX3CR1 sense 5’- gaattcaccatggatcagttccctgaatcagt -3’

hCX3CR1 antisense 5’- ctcgagtcagagaaggagcaatgcatctccatca -3’

hCX3CR1 S319X antisense 5’- cggtgaagaatatgcttccaaaaaagccgta -3’

hCX3CR1 R127N sense 5’- tcatcagcattgataactacctggccatcgtcct -3’

hCX3CR1 R127N antisense 5’- cggtgatgaagaatatgcttccaaaaaagccgta -3’

hCX3CR1 N289A sense 5’- cctggctcctctcatctatgcattt -3’

hCX3CR1 N289A and Y293A antisense 5’- caacaatggctaaatgcaaccgtct -3’

hCX3CR1 Y293A sense 5’- cctgaatcctctcatcgctgcattt -3’

hCXCR6 sense 5’- gaattcaccatggcagagcatt-3’

hCXCR6 antisense 5’-ctcgagctataactggaacat-3’

hCXCR6 R127N sense 5’-cactgtggatcgttacattgta-3’


26

hCXCR6 R127N antisense 5’-atgcaggtgaggatgagcat-3’

hCXCR6 F128Y sense 5’-atgcaggtgaggatgagcat-3’

hCXCR6 F128Y antisense 5’-atgcaggtgaggatgagcat-3’

4.1.3 Plasmids

Name Backbone Restriction sites used

hCX3CR1 pcDNA3.1+ EcoRI XhoI

hCX3CR1-R127N pcDNA3.1+ EcoRI XhoI

hCX3CR1-N289A pcDNA3.1+ EcoRI XhoI

hCX3CR1-Y293A pcDNA3.1+ EcoRI XhoI

hCX3CR1-S319X pcDNA3.1+ EcoRI XhoI

hCXCR6 pcDNA3.1+ EcoRI XhoI

hCXCR6-R127N pcDNA3.1+ EcoRI XhoI

hCXCR6-F128Y pcDNA3.1+ EcoRI XhoI

4.1.4 Antibodies

Antibody Species Company Final Conc. Application

hCX3CR1-PE rat IgG2bκ MBL, Japan 1 µg/ml FACS

isotype control-PE rat IgG2bκ Abcam 1 µg/ml FACS

hCX3CL1 mouse IgG1 R&D Systems 10 µg/ml FACS, inhibition

hCX3CL1 mouse R&D Systems 4 µg/ml ELISA

hCX3CL1-biotin mouse R&D Systems 0.3 µg/ml ELISA

ICAM-1 mouse IgG1 R&D Systems 10 µg/ml inhibition

VCAM-1 mouse IgG1 R&D Systems 10 µg/ml inhibition


27

isotype control mouse IgG1 R&D Systems 10 µg/ml FACS, inhibition

hCXCR6-PE mouse IgG2b R&D Systems 5 µg/ml FACS

isotype control-PE mouse IgG2b R&D Systems 5 µg/ml FACS

6xHis-tag mouse IgG2b Abcam 4 µg/ml FACS

mouse-PE goat Jackson Immuno 2.5 µg/ml FACS

humanFc-PE goat Jackson Immuno 5 µg/ml FACS

4.1.5 Chemicals

All chemicals were ordered from either Sigma-Aldrich (St.Louis, MO) or Roth

(Karlsruhe, Germany), if not specified otherwise. All enzymes for cloning were

purchased from Fermentas (St.Leon-Rot, Germany), if not specified otherwise.

Name Company

AlexaFluor647-labeled CX3CL1-CD Almac, Craigavon, UK

AMAXA nucleofactor Lonza, Cologne, Germany

Calcein-AM Biotium Inc., Hayward, CA

Collagen G Biochrom AG, Berlin, Germany

Collagenase II Sigma-Aldrich, St.Louis, MO

CXCL16-CD Peprotech, Hamburg, Germany

CXCL16-6xHis R&D Systems, Wiesbaden, Germany

CX3CL1-CD Peprotech, Hamburg, Germany

Endothelial cell growth medium Promocell, Heidelberg, Germany

Fluo-3-AM Molecular Probes, Karlsruhe, Germany

Gel loading dye Fermentas, St.Leon-Rot, Germany

G418 Calbiochem, Hamburg, Germany

HBSS Gibco, Karlsruhe, Germany


28

HEPES Gibco, Karlsruhe, Germany

IFNγ Peprotech, Hamburg, Germany

Lipofectamine Invitrogen, Karlsruhe, Germany

NucleoSpin extract II kit Macherey-Nagel, Dueren, Germany

Pancoll Pan Biotech, Aidenbach, Germany

pcDNA3.1+ Invitrogen, Karlsruhe, Germany

Pertussis toxin Calbiochem, Hamburg, Germany

Plasmid DNA purification kit Macherey-Nagel, Dueren, Germany

Strep-POD Roche, Grenzach, Germany

TNFα Peprotech, Hamburg, Germany

TOPO TA cloning kit Invitrogen, Karlsruhe, Germany

4.1.6 Devices and Materials

Device/Material Company

AMAXA Nucleofactor Lonza, Cologne, Germany

Avanti J-25 Beckman Coulter, Palo Alto, CA

Boyden chamber Neuroprobe, Gaithersburg, MD

Cary Eclipse reader Varian, Palo Alto, CA

Centrifuge 5415R Eppendorf, Hamburg, Germany

Centrifuge 5810R Eppendorf, Hamburg, Germany

EasyCyte Mini Millipore, Billerica, MA

FACSCalibur BD Biosciences, San Jose, CA

FACSCanto BD Biosciences, San Jose, CA

Fluorescence plate reader Tecan Genios Tecan, Maennedorf, Switzerland


29

Glass bottom dishes MatTek, Ashland, MA

Hemacytometer Brand, Wertheim, Germany

Leica DM-IL microscope Leica, Wetzlar, Germany

Leica DM-IRB microscope Leica, Wetzlar, Germany

LSM 7 Duo microscope Zeiss, Goettingen, Germany

MaxiSorp ELISA plate Nunc, Langenselbold, Germany

µ-slide VI ibidi, Munich, Germany

NanoDrop ND-1000 PEQLAB, Erlangen, Germany

Polycarbonate membrane Neuroprobe, Gaithersburg, MD

Syringe pump Landgraf, Langenhagen, Germany

TProfessional basic thermocycler Biometra, Goettingen, Germany

Transwell filters Costar Corning, Schiphol-Rijk, Netherlands

4.1.7 Inhibitors

The substance GW280264X is a metalloprotease inhibitor based on hydroxamat

(Figure 5). It was synthesized at GlaxoSmithKline (Stevenage, UK) and stored at 4°C in

a 10 mM stock in DMSO. GW280264 inhibits the proteolytic activity of ADAM17, and

to a smaller degree of ADAM10 [Ludwig et al., 2005].

Figure 5: Structure of GW280264X. The metalloprotease inhibitor is based onhydroxamat and inhibits ADAM17 as well as ADAM10. The half maximal in-hibitory concentration (IC50) is 8.0 nM for ADAM17 and 11.5 nM for ADAM10.


30

Dynasore (3-Hydroxynaphthalene-2-carboxylic acid (3,4-dihydroxybenzylidene)

hydra-zide) was obtained from Tocris (Bristol, UK) and stored at -20°C at 100 mM in

DMSO. The working concentration was 100 µM. It is a small molecule GTPase inhibitor

that targets dynamin-1 and dynamin-2. Dynasore blocks dynamin-dependent

endocytosis, and scission of endocytic vesicle. It is cell-permable and blocks in a

reversible and noncompetitive manner [Macia et al., 2006].

Figure 6: Structure of Dynasore. The small molecule GTPase inhibitor Dynasoreblocks dynamin-1 and dynamin-2. The half maximal inhibitory concentration(IC50) is approximately 15 µM.

4.2 Methods

4.2.1 Cell culture

4.2.1.1 Maintenance of cells

The human embryonic kidney cell line HEK293 (ATCC: CRL-1573) was grown in

DMEM medium supplemented with 1% penicillin and streptomycin and 10% FCS. The

human epithelial cell line ECV304 (ATCC: CRL-1998) was grown in M199 medium

supplemented with 1% penicillin and streptomycin and 10% FCS. The murine pre-B-cell

cell line L1.2 (Massachusetts Association of Technology Transfer, Tufts University,

Boston, MA) was grown in RPMI1640 medium supplemented with gentamycin, 10%

FCS, 1% HEPES, 1% pyruvate and β-mercaptoethanol. The primary human umbilical


31

vein endothelial cells were grown in Endothelial Cell Growth Medium including

Supplement.

Cells were subcultured using trypsin-EDTA. To this purpose, cells were washed with

PBS and incubated with Trypsin-EDTA for 5 minutes at 37°C. Then, cells were

resuspended in PBS, centrifuged at 300×g, and seeded at 1:5 to 1:20.

Cells were counted and viability of cells was tested using Trypan blue. 10 µl of cell

suspension was mixed with 10 µl 0.4% Trypan blue and given onto a hematocymeter.

Trypan blue exclusively stains dead cells blue. These cells were excluded in counting.

For freezing cells, confluent grown cells were washed with PBS and incubated with

Trypsin-EDTA for 5 minutes at 37°C. Subsequently, cells were washed with PBS,

centrifuged at 300×g at room temperature, and resuspended in 3 ml pre-cooled medium

with 20% FCS and 10% DMSO. 1 ml of cell suspension at a time was given into a

cryotube. Cells were then transferred on ice at -20°C over night, at -80°C for 3 days, and

at liquid nitrogen for final storage.

4.2.1.2 Transient transfection

Different strategies were used to transiently transfect cells. HEK293 cells were

transfected with lipofectamine, while L1.2 cells were transfected using the AMAXA

system.

Lipofectamine. Cells were seeded in 6-well-plates and grown to 70% - 80%

confluency. Following complexes were prepared for transfection per well of a 6-well-

plate:

1.5 µg DNA

3 µl Lipofectamine

100 µl serum-free medium


32

During the incubation of the reaction setup for 20 minutes at room temperature, cells

were washed and medium was exchanged. 100 µl of the complex solution were added to

each well drop by drop. Cells were incubated at 37°C in a CO2 incubator for 48 hours.

AMAXA Nucleofactor. 2x106 cells were taken out of the flask, centrifuged at 300×g

for 5 minutes at room temperature and medium was removed. The cell pellet was

resuspended with the following solution:

5 µg DNA

18.1 µl Nucleofactor supplement

81.8 µl Nucleofactor solution

The cell/DNA suspension was then transferred into cuvettes, inserted into the

Nucleofactor device, and program P16 was applied. Immediately after the program was

finished, pre-warmed medium was added to the cuvettes, and the cells were gently

transferred into T25 flasks. Cells were incubated at 37°C in a CO2 incubator for 48 hours.

4.2.1.3 Stable transfection

48 hours after transfection cells were selected for receptor expression using 1 mg/ml

G418 because only transfected cells are resistent to G418. For HEK293 cells medium

was changed every 3 days, and fresh G418 was added. L1.2 cells were tested for viability

every 3 days, and fresh G418 was added every 5 days. The remaining viable cells were

cultured to confluency, seeded on 10 cm2 plates and partly frozen when confluent. Single

clones were established using the limited dilution method. Cells were counted and

diluted to 1 cell/100 µl medium containing G418. 100 µl of cell suspension was added

per well of a 96-well-plate. After 5 days, 50 µl medium containing 3x G418 was added to

the wells. Formation of single cell patches was controlled by microscopy. After reaching

confluency, cells were gently transferred to 6-well-plates and T25 flasks. The receptor

expression level was checked using flow cytometry analysis. Clones of the different


33

receptor variants with similar expression levels were chosen for further experiments.

4.2.1.4 Isolation of HUVEC

Primary human umbilical vein endothelial cells were freshly prepared from human

umbilical veins. For this purpose, 10 - 15 cm of umbilical vein were flushed with PBS to

avoid contamination with blood cells. Afterwards, the vein was filled with collagenase

type II solution (40 µg/ml in PBS) and incubated for 15 minutes at 37°C. Cells were then

flushed out of the vein with 50 ml PBS and centrifuged at 300xg at room temperature.

Cells were resuspended in supplemented HUVEC growth medium and grown on

collagen G-coated cell culture dishes (coating: 15 minutes with 40 µg/ml collagen G in

PBS). HUVEC were subcultured before reaching confluency for a maximum of 4

passages.

4.2.1.5 Isolation of PBMCs

Human peripheral blood mononuclear cells (PBMC) were isolated from citrated

(0.38%) venous blood of healthy volunteers by 1:1 dilution in PBS and subsequent

density gradient centrifugation on a Ficoll Hypaque layer (25 ml blood-PBS mixture on

25 ml Ficoll Hypaque) for 40 minutes at RT without brake. PBMC that accumulated in

the middle layer were carefully taken off and, after 2-fold washing with PBS, PBMC

were used for functional assays as decribed below.

4.2.2 Cloning

4.2.2.1 hCX3CR1 and its variants in pcDNA3.1

Human CX3CR1 cDNA was previously amplified by Matthias Voss from human


34

PBMC using the primers indicated in section 4.1.2 and cloned into pcDNA3.1+ using

XhoI and EcoRI as restriction sites generating hCX3CR1-pcDNA3.1+. Based on this

plasmid, the truncation variant of hCX3CR1 that lacks all intracellular serine-residues

was generated by using an antisense primer with inserted stop codon at serine residue

319 and subsequent restriction site for XhoI. cDNA was amplified, and the product was

purified using gelelectrophoresis and a gel extraction kit. cDNA for the other receptor

variants was generated by site-directed mutagenesis via ligation PCR. An upstream

fragment was amplified using the CX3CR1 sense primer and a specific antisense primer

for the hCX3CR1 variant. Additionally, a downstream fragment was generated with

antisense primer for CX3CR1 and sense primers carrying the desired mutations. The

upstream and downstream PCR products were phosphorylated separately using

polynucleotide kinase and ligated before a PCR with CX3CR1 sense and antisense

primers was performed. The cDNA for the different receptor variants was ligated into

TOPO cloning vector and positive clones were selected with blue-white screening.

Clones were then grown and DNA was purified. The product was subsequently ligated

into the expression vector pcDNA3.1+ using EcoRI and XhoI. All sequences were

confirmed by sequencing. All plasmids were amplified in large scale and frozen for

immediate use. Protocols for the different steps are given below.

Human CXCR6 cDNA was previously generated by Alexander Schulte from human

PBMC using the primers indicated in section 4.1.2 and cloned into pcDNA3.1+ using

XhoI and EcoRI as restriction sites generating hCXCR6-pcDNA3.1+ [Ludwig et al.,

2005]. Based on this plasmid, the variants of hCXCR6 were generated by the same

strategy used for the CX3CR1 variants.

PCR. For the variants of human CXCR6 and CX3CR1, ligation-PCRs were

performed. Alteration of the DRY and NPX2-3Y motifs required two PCRs leading to an

upstream and a downstream fragment. DNA polymerase of Pyrococcus furiosus (Pfu

polymerase) was used due to its integrated proof reading function and the generation of

blunt end fragments. The following protocol was used:


35

1 µl forward primer

1 µl reverse primer 10' 95°C initial denaturation

0.5 µl pfu polymerase 2' 95°C denaturation

5 µl pfu buffer 35 cycles 30'' 60°C primer annealing

2 µl dNTP 2' 72°C elongation

1 µl plasmid 10' 72°C final elongation

39.5 µl A.dest.

Agarose gel electrophoresis. After PCR was performed, products were separated

using agarose gel electrophoresis. Therefore, samples were mixed with 2x loading dye,

loaded onto 1% or 2% agarose gels including ethidiumbromide in TAE buffer and

separated using 100 V. Afterwards, bands of the desired size were cut on an UV-light

table. As standard, 5 µl of GeneRuler kb DNA Ladder was used.

TAE buffer (50x)

242 g Tris

57.1 ml acetic acid

100 ml 0.5 M EDTA, pH 8.0

ad 1000 ml A.dest.

Gel extraction. For DNA gel-extraction, the NucleoSpin extract II kit was used

according to manufacturer’s instructions. Briefly, gel fragment was mixed with buffer at

0.5 mg/µl and bound onto a column. After washing off salt and impurities, DNA was

eluted in 50 µl H2O.

Phosphorylation of PCR fragments. PCR fragments had to be phosphorylated for

ligation. The following mixture was prepared for each fragment:


36

10 µl 10 x ligation buffer

1 µl PCR product

→ 5' 70°C

2 µl T4 polynucleotide kinase

→ 30' 37°C

Ligation of PCR fragments. Subsequently, both phosphorylated fragments were

ligated using T4 ligase in the following mixture:

5 µl phosphorylated upstream fragment

5 µl phosphorylated downstream fragment

1 µl T4 DNA ligase

→ 60' RT

PCR of ligation product. Ligation products were then amplified using PCR

technique. In this case, high fidelity Taq polymerase was used due to its integrated proof

reading function and the generated A-overhang in products. This PCR protocol was also

used for the truncation mutant of the receptor.

1 µl forward primer

1 µl reverse primer 10' 95°C initial denaturation

0.5 µl high fidelity Taq polymerase 2' 95°C denaturation

5 µl high fid. Taq (incl Mg2+) buffer 35 cycles 30'' 60°C primer annealing

2 µl dNTP 2' 72°C elongation

1 µl ligation or plasmid 10' 72°C final elongation

39.5 µl A.dest.

Subsequently, PCR products were separated using agarose gel electrophoresis and the

fragment at about 1 kb was cut and extracted.


37

TOPO TA cloning. For further amplification PCR products were cloned into TOPO

vector using the TOPO TA cloning kit according to manufacturer’s instructions.

Therefore, the following mixture was prepared:

4 µl PCR product

1 µl salt solution

1 µl vector

→ 10' RT

This mixture was then used for heat-shock transformation.

Heat-shock transformation. To transfer the plasmid containing the desired sequence

into bacteria, heat-shock transformation was used. E.coli TOP10 bacteria were incubated

with 1.5 µl of plasmid for 30 minutes on ice. For heat-shock, cells were incubated for 40

seconds at 42°C and then incubated for 5 minutes on ice. Cells were then mixed with 450

µl SOC medium (SOB medium supplemented with 20 mM glucose) and incubated for 1

hour at 37°C while shaking. 100 µl of cell suspension were spread on agar plates

containing ampicillin and X-gal for TOPO cloning and only ampicillin for pcDNA3.1+

cloning. Plates were incubated over night at 37°C.

Clone selection. For TOPO cloning, colonies were selected via blue-white-screening.

White colonies represent positive clones that integrated the plasmid, whereas blue

colonies did not include the plasmid. For pcDNA3.1+ cloning, colonies were picked

randomly with a pipette tip and tested for transformation via colony PCR technique.

Picked clones were transferred into 5 ml LB medium and grown over night at 37°C.

Colony PCR. Successful transformation was tested via PCR technique. As template

the pipette tip with the picked colony was dipped into the reaction mix. Following

protocol was used:


38

0.5 µl forward primer

0.5 µl reverse primer 10' 95°C initial denaturation

0.5 µl Taq polymerase 1' 95°C denaturation

3 µl Taq (excl. Mg2+) buffer 30 cycles 30'' 63°C primer annealing

5 µl Mg2+ 1' 72°C elongation

1 µl dNTP 10' 72°C final elongation

39.5 µl A.dest.

Subsequently, the products were visualized using agarose gel electrophoresis.

MiniPrep. Amplified plasmids were purified using a plasmid DNA purification kit

according to manufacturer’s instruction. Briefly, cells were sedimented, lysed and DNA

and proteins were precipitated. DNA was then bound onto a silica column. After washing

off contaminants, plasmids were eluted in A.dest.

Restriction. In order to express the gene product in mammalian cells, the cDNA was

transferred from the TOPO vector into the eukaryotic expression vector pcDNA3.1+.

Therefore, the desired sequence was cut out of the TOPO vector, and the pcDNA3.1+

vector was cut for linearization using the following mixture:

1 µl EcoRI

1 µl XhoI

6 µl Tango Buffer

1 µl plasmid (TOPO and pcDNA3.1+)

21 µl A.dest.

→ 3 h 37°C

Subsequently, products were separated using agarose gel electrophoresis, and the

fragment at about 1 kb and the linearized pcDNA3.1+ vector were cut and extracted.


39

Ligation into pcDNA3.1+. For ligation of the DNA into pcDNA3.1+ with T4 ligase

the following protocol was used:

10 µl insert

1 µl plasmid

1 µl T4 DNA ligase

3 µl T4 DNA ligase buffer

15 µl A.dest.

→ over night 16°C

The ligated plasmid was then transferred into E.coli using heat-shock transformation

as described above.

4.2.2.2 Fractalkine-Fc fusion protein

An expression vector for the fusion protein of human CX3CL1 and human IgG1-Fc-

fragment (hFc) was generated previously by Matthias Voss. cDNA coding for hFc was

inserted into pcDNA3.1 using EcoRI and XhoI. Subsequently, a cDNA fragment for the

ectodomain of human CX3CL1 was generated by PCR using sense and antisense primers

as indicated in table 4.1.2, and then inserted into pcDNA3.1-Fc using HindIII and EcoRI.

All sequences were confirmed by sequencing.

4.2.3 Molecular biology and proteinbiochemistry

4.2.3.1 Analysis of surface expression

Cells were analyzed for surface expression of chemokine receptors or transmembrane

chemokines by staining with appropriate antibodies (see table 4.1.4). Cells were


40

harvested, washed with PBS, and resuspended to 1x106 cells/ml in PBS with 0.2% BSA.

Cells were then incubated with antibody solution at the indicated final concentration and

in a final volume of 25 µl. For directly fluorescently labeled antibodies, cells were

stained for 30 minutes on ice in the dark, washed twice with 250 µl PBS with 0.2% BSA

to remove excess antibody, and resuspended in 100 µl PBS with 0.2% BSA. For

unlabeled antibodies, cells were incubated with first antibody on ice for 30 minutes, after

washing they were incubated with the fluorescently labeled detection antibody against

the first antibody on ice in the dark for another 30 minutes and washed afterwards. For

immediate measurements, no fixation was used, and cells were resuspended in PBS with

0.2% BSA, otherwise cells were resuspended in 4% PFA in PBS with 0.2% BSA and

kept cold in the dark. The fluorescence signal was then analyzed by flow cytometry.

PBS (phosphate buffered saline)

137 mM NaCl

2.7 mM KCl

10 mM Na2HPO4

2 mM KH2PO4

in A.dest.

4.2.3.2 ELISA

For ELISA (enzyme-linked immunosorbent assay) detection of soluble CX3CL1,

MaxiSorp flat-bottom 96-well plates were coated over night with coating antibody at a

final concentration of 4 µg/ml in PBS at room temperature. Plates were then washed 3

times with washing buffer (PBS with 0.05% Tween) and blocked for 2 hours with

blocking buffer (PBS with 0.05% Tween and 2% BSA) at room temperature. Samples

and standard were added after washing and incubated for 1 hour at room temperature.

After washing 3 times with washing buffer, anti-CX3CL1-biotin antibody in a final


41

concentration of 0.3 µg/ml in blocking buffer was added and incubated for 1 hour at

room temperature. After washing with washing buffer, Streptavidin-POD was added in a

1:5000 dilution in blocking buffer, and incubated for 1 hour at room temperature. After

washing 4 times with washing buffer, substrate solution was added, incubated for 10-20

minutes and the reaction was stopped using 100 µl 1M H2SO4. The signal was then

measured in a plate reader at a wavelength of 450 nm with a reference wavelength of 550

nm.

4.2.4 Functional assays

4.2.4.1 Ligand binding

For CX3CL1 binding experiments, cells were harvested, washed with PBS and

resuspended to 2x106 cells/ml in PBS with 0.2% BSA. Cells were then incubated for 1

hour on ice with a 1:10 dilution of the supernatant of HEK293 cells transfected to

express soluble CX3CL1 that was fused to the Fc-part of human IgG1. After washing off

excess CX3CL1-Fc twice with 250 µl PBS with 0.2% BSA, the Fc-part was detected

using a fluorescently labeled α-Fc-antibody in a final volume of 25 µl. Cells were

incubated with the antibody for 30 minutes on ice in the dark. After washing off excess

antibody twice with 250 µl PBS with 0.2% BSA, cells were resuspended in 100 µl PBS

with 0.2% BSA and the fluorescence signal was measured by flow cytometry.

For CXCL16 binding experiments, cells were harvested, washed with PBS and

resuspended to 2x106 cells/ml in PBS with 0.2% BSA. Cells were then incubated for 1

hour on ice with 100 ng/ml recombinant 6xHis-tagged CXCL16 in the absence or

presence of 1 µg/ml recombinant untagged CXCL16. After washing off excess CXCL16

twice with 250 µl PBS with 0.2% BSA, the 6xHis-tag was detected with a mouse

anti-6xHis-antibody for 30 minutes on ice in a final volume of 25 µl. After two washing

steps with 250 µl PBS with 0.2% BSA, cells were incubated with a goat anti-mouse PE-


42

labeled antibody for 30 minutes on ice in the dark. After washing off excess antibody

twice with 250 µl PBS with 0.2% BSA, cells were resuspended in 100 µl PBS with 0.2%

BSA and the fluorescence signal was measured by flow cytometry.

4.2.4.2 Ligand uptake

For ligand uptake experiments, cells were harvested and resuspended to 2x106 cells/ml

in PBS with 0.2% BSA. Prior to CX3CL1-treatment, cells were left on ice for 30 minutes

with or without 0.2% NaN3. Cells were then incubated with 10 nM AlexaFluor647-

labeled CX3CL1 chemokine domain at 4°C or 37°C for 30 minutes in a final volume of

25 µl. Cells were washed twice with 250 µl ice-cold PBS with 0.2% BSA or PBS with

0.2% BSA containing 0.2% NaN3, resuspended with ice-cold PBS with 0.2% BSA or

PBS with 0.2% BSA containing 0.2% NaN3 and kept on ice to avoid further ligand

uptake. The fluorescence signal was measured by flow cytometry.

4.2.4.3 Intracellular calcium transients

Cells were harvested, resuspended in PBS to 2x106 cells/ml, and loaded with 4 µg/ml

of the calcium-indicator Fluo-3-AM for 30 minutes at 37°C. After two-fold washing with

calcium assay buffer, cells were resuspended and kept in calcium assay buffer

supplemented with 1% FCS. Before measurement, 2.5x106 cells were resuspended in 1.5

ml assay buffer containing no calcium, but 0.3 mM EDTA. Cells were incubated in a

plastic cuvette at 37°C under constant stirring and Fluo-3 fluorescence was continuously

monitored at excitation and emission wavelengths of 490 and 526 nm, respectively, using

a fluorescence spectrophotometer. After 100 seconds, cells were stimulated with 10 nM

CX3CL1 or CXCL16 chemokine domain, and the fluorescence signal was recorded for

200 seconds. Finally, loading of cells with fluorescent dye was controlled by addition of

calcium (1.6 mM) and digitonin (125 µg/ml), leading to maximal signal by complex

formation of Fluo-3 with calcium, and finally by addition of EGTA (15 mM) for


43

dissolving the complex.

Calcium Assay Buffer

138 mM NaCl

6 mM KCl

1 mM MgCl2

5.5 mM D-Glucose

20 mM HEPES

1.6 mM CaCl2

in A.dest.

pH 7.4

4.2.4.4 Adhesion under static conditions

For adhesion under static conditions, cells expressing the ligand (HUVEC or ECV304)

were seeded onto 24-well-plates and were cultured to full confluency as confirmed by

microscopy. HUVEC were pretreated with 10 ng/ml TNFα and 10 ng/ml IFNγ 16 hours

prior to the experiment to stimulate chemokine expression. L1.2 cells or PBMC were

harvested and stained with 10 µM CalceinAM in PBS for 15 minutes at 37°C in the dark,

and after washing off excess dye, 5x105 cells in PBS were centrifuged with 300xg for 3

minutes onto the chemokine expressing HUVEC or ECV304 cell layer. The fluorescence

signal of adherent cells was measured in a fluorescence plate reader after each of 12

washing steps with PBS at an excitation wavelength of 480 nm and an emission

wavelength of 550 nm.

The number of adherent cells is given as adhesion index (AI) that was determined by

dividing the number of adherent cells on ligand-expressing cells by random adhesion of


44

cells on a cell layer that does not express the ligand.

4.2.4.5 Adhesion under flow conditions

For adhesion under flow conditions, ECV304 cells or HUVEC were seeded onto a

flow adhesion chamber and were cultured to full confluency. HUVEC were stimulated

with 10 ng/ml IFNγ and TNFα 16 hours prior to assay to induce CX3CL1 expression.

HEK293 cells or PBMC were stained with 10 µM CalceinAM for 15 minutes at 37°C in

the dark, washed to remove of excess dye and resuspended in flow adhesion buffer to

5x105 cells/ml. The flow adhesion chamber containing the ECV304 or HUVEC cell layer

was then connected to a syringe pump and inserted into a temperature controled

incubation chamber culture on the stage of an inverted microscope. After washing the

cell layer with warm flow adhesion buffer for 3 minutes, it was perfused by a suspension

of calcein-labeled cells with a shear stress rate of 2 dyne for 1 minute. Thereafter, 8

subsequent pictures were taken, and the number of adherent cells was determined.

Flow Adhesion Assay Buffer

10% HBSS

1% HEPES

1% BSA

in A.dest.

The number of adherent cells is given as adhesion index (AI) that was determined by

dividing the number of adherent cells on ligand-expressing cells by random adhesion of

cells on a cell layer that does not express the ligand.

4.2.4.6 Chemotaxis

Stably transfected L1.2 cells or PBMC were harvested, washed twice with RPMI1640


45

with 0.2% BSA, and resuspended at 2x106 cells/ml in RPMI1640 with 0.2% BSA. The

lower wells of a modified 48-well Boyden chamber were filled with 29.5 µl RPMI1640

with 0.2% BSA with or without the indicated concentrations of CX3CL1 or CXCL16

chemokine domain, covered with a polycarbonate membrane with 8 µm pores, and 30 µl

of cell suspension was added to the upper wells. After incubation of 2 hours at 37°C,

migrated cells were counted using a hemacytometer.

The number of migrated cells is given as migration index (MI) that was determined by

dividing the number of migrated cells treated with a stimulus by random migration of

unstimulated cells.

4.2.4.7 Transmigration

ECV304 cells or HUVEC were seeded on transwell filters with 8 µm pores and grown

to confluency. L1.2 cells or PBMC were harvested, washed twice with with 0.2% BSA,

and resuspended to 2x106 cells/ml in RPMI1640 with 0.2% BSA. 24-well culture plates

were filled with 600 µl/well RPMI1640 with 0.2% BSA with or without the indicated

concentrations of CX3CL1 or CXCL16 chemokine domain. Transwell inserts containing

the HUVEC or ECV304 cell layer were then placed into the wells and filled with 100 µl

of L1.2 or PBMC cell suspension. After incubation for 2 hours at 37°C, transmigrated

cells were counted using a hemacytometer.

The number of transmigrated cells is given as transmigration index (TI) that was

determined by dividing the number of transmigrated cells treated with a stimulus by

random transmigration of unstimulated cells.

4.2.4.8 Pseudopod formation

Wt- or CX3CL1-ECV304 cells were seeded on glass bottom dishes and grown to

confluency. L1.2 cells were harvested, washed twice with RPMI1640 with 0.2% BSA,


46

resuspended to 2x105 cells/ml in RPMI1640 with 0.2% BSA and placed on the ECV304-

cell layer in the prewarmed chamber at 37°C. Time-lapse videos were captured using a

LSM 7 Duo Microscope for 5 minutes, and 100 cells per condition were analyzed for

pseudopod formation. Evaluation for pseudopod formation defined as temporal

protrusions from the cell body with a minimal length of 0.5 µm was done in a blinded

fashion.

4.2.5 Statistical analysis

Data were statistically analyzed by using the one-way ANOVA with posthoc

Bonferroni’s Multiple Comparison t-tests using GraphPadPrism software (GraphPad

Prism 5.01, GraphPad Software, San Diego, CA). In case of heteroscedasticity (Bartlett

test), data were log transformed prior to analysis. The data in Figures 23 and 25 were

analyzed by the Mixed Model procedure in SAS 9.1 (SAS Institute, Cary, NC). Multiple

comparisons were corrected by the Bonferroni-Holm procedure. In the figures only those

comparisons with interest to the main hypotheses are indicated.


47

5 Results

5.1 CX3CR1-CX3CL1 interaction

In order to characterize the interaction between CX3CR1 and CX3CL1, I started by

determining whether the interaction of fractalkine and CX3CR1 results in firm adhesion

and (trans-)migration of PBMC. Since CX3CR1 is described to be constitutively

expressed on monocytes and T-cell populations, peripheral blood mononuclear cells

from blood of healthy donors were isolated, and tested for the expression of CX3CR1

using a PE-labeled antibody in flow cytometry. I found that the level of expression

varied depending on donor and cell type. As expected, the highest expression of

CX3CR1 could be seen on monocytes, and lower expression was found on a

subpopulation of T-cells (Figure 7). For further experiments PBMC from donors with a

strong expression of CX3CR1 were used.

Isotype

CX3CR1

Fluorescence Intensity

Rela

tive C

ell

Num

ber

Figure 7: CX3CR1 is expressed on PBMC. Freshly isolated blood mononuclearcells of healthy donors were probed for their expression of CX3CR1 using a PE-la-beled anti-hCX3CR1 antibody. Cells were gated for monocytes by sideward andforward scatter dotplot. The grey shaded curve represents the isotype control, theblack unshaded curve represents hCX3CR1 expression. Data are shown as a repre-sentative histogram.

5 Results

48

A fundamental step in leukocyte recruitment is the firm adhesion on the endothelium

lining the blood vessel. Therefore, freshly isolated PBMC were tested for their ability to

mediate adhesion under flow conditions. For this purpose, wt-ECV304 or CX3CL1-

ECV304 were grown to confluency to form a stable cell layer to which PBMC could

adhere. Using a syringe pump, freshly isolated PBMC were directed over this cell layer

for 1 minute with a shear force stress similar to that of the post-capillary system (2 dyne/

cm2), and the adherent PBMC were counted. 8.3 times more PBMC adhered on a

CX3CL1-expressing ECV304 cell layer compared to a wt-ECV304 cell layer (Figure 8),

indicating that CX3CL1 may function as adhesion molecule as was shown earlier.

!"#$%&'() %*'%+,#$%&'()(-(

.-/

/-(

0-/

,(-( 1

23456789:;935<

Figure 8: PBMC adhere to a CX3CL1 expressing cell layer under flow condi-tions. Freshly prepared PBMC were assayed for adhesion under flow conditions ona wt-ECV304 or a CX3CL1-ECV304 cell layer. PBMC were stained with Cal-ceinAM for 15 minutes. Excess dye was removed by washing, and cells were sus-pended to a concentration of 5x105 cells/ml in warmed flow adhesion buffer. TheECV304 cell layer was washed with warmed flow adhesion buffer, and then PBMCwere directed onto the layer under a steady shear stress rate of 2 dyne/cm2 under afluorescence microscope. Adherent cells were counted after 1 minute. Data areshown as mean ± SD of three independent experiments. *, p<0.05 versus the un-transfected control.

Since leukocyte adhesion is usually followed by transmigration through endothelial

cells, I tested PBMC for their ability to migrate towards CX3CL1. Freshly prepared

5 Results

49

PBMC were assayed for chemotaxis in response to soluble CX3CL1 chemokine domain

(1 - 100 nM) in a modified 48-well Boyden chamber setting. 2.2 times more PBMC

migrated into the lower well when 10 nM CX3CL1 was present in the lower well.

Chemotaxis towards CX3CL1 was dose-dependent and showed the characteristic bell

shaped curve reported for chemokines, peaking at 10 nM soluble CX3CL1 chemokine

domain (Figure 9). Based on this curve, all further experiments were performed with 10

nM soluble CX3CL1 chemokine domain.

00

1

2

3

0.1 1 10 100

*

* *

CX3CL1 Concentration [nM]

Mig

ratio

n In

dex

Figure 9: PBMC chemotactically migrate towards CX3CL1. PBMC were seed-ed onto a polycarbonate membrane with 8 µm pores in a modified Boyden chambersetting. 1 - 100 nM soluble CX3CL1 chemokine domain were used to excitechemotaxis for 2 hours at 37°C. Subsequently, migrated cells were counted. Dataare shown as mean ± SD of three independent experiments. *, p<0.05 versus ran-dom migration without stimulus.

Chemotaxis describes the process of cell movement through a membrane instead of

the interaction of different cell types with each other and subsequent movement through

a confluent cell layer (transmigration). To study the latter process, the capability of

freshly isolated PBMC to transmigrate through a cell layer expressing CX3CL1 was

tested. Transmigration was studied in the absence or presence of soluble CX3CL1 that

was added to the lower compartment of a transwell insert. As expected, transmigration of

5 Results

50

PBMC was 1.9 fold increased when soluble CX3CL1 was used as a chemoattractant.

Transmigration was also increased (Transmigration Index (TI) = 1.7) when

transmembrane CX3CL1 was expressed on the ECV304 cell layer. However,

simultaneous presence of transmembrane and soluble CX3CL1 did not further enhance

transmigration (TI = 1.8) (Figure 10).

wt-ECV304 CX3CL1-ECV3040.0

0.5

1.0

1.5

2.0

2.5none10nM CX3CL1

Tran

smig

ratio

n In

dex *

Figure 10: PBMC chemotactically transmigrate in response to CX3CL1. Fresh-ly prepared PBMC were assayed for transmigration through wt- or CX3CL1-ECV304 cell layers cultured in transwell inserts. After washing, 2x105 PBMC wereseeded onto the confluent cell layer for 2 hours at 37°C. Subsequently, transmigrat-ed cells were counted. Data are shown as mean ± SD of three independent experi-ments. *, p<0.05 versus the migration without the addition of stimulus.

In order to use primary endothelial cells as cell layers in transmigration assays,

HUVEC were isolated from human umbilical veins. Under normal cell culture

conditions, they do not express CX3CL1 endogenously. When these cells were

stimulated with IFNγ and TNFα for 16 hours, they expressed CX3CL1 in considerable

amounts, as was shown earlier and by FACS analysis (Figure 11) [Ludwig et al., 2002].

5 Results

51

Figure 11: Stimulated HUVEC express CX3CL1. Primary human umbilical veinendothelial cells were stimulated with IFNγ and TNFα (both 10 ng/ml) for 16 hoursor left untreated, and were subsequently stained for CX3CL1 surface expression us-ing an anti-CX3CL1 antibody. The grey shaded curve represents the unstimulatedcontrol, the black unshaded curve represents hCX3CL1 expression after IFNγ andTNFα treatment. Data are shown as a representative histogram.

To analyze transmigration in a more physiologic setting, I next used HUVEC as a cell

layer for transmigration. Transmigration of PBMC was considerably increased (TI = 3)

(Figure 12) when these cells were stimulated with IFNγ and TNFα to express

endogenous CX3CL1. This transmigration was almost completely suppressed by

pretreatment of the HUVEC layer with a neutralizing antibody to CX3CL1. Since firm

adhesion occurs prior to the transmigration process, adhesion molecules like ICAM-1 or

VCAM-1 might also play an important role in transmigration of PBMC towards

CX3CL1. Additionally, treatment of HUVEC with IFNγ and TNFα might not only

increase CX3CL1 expression, but also that of adhesion molecules. To find out whether

CX3CL1 is sufficient to promote the transmigration process, neutralizing antibodies to

ICAM-1 and VCAM-1 were used. The inhibition of VCAM-1 and ICAM-1 was as

efficient as that obtained with neutralizing antibodies to CX3CL1 (TI = VCAM-1: 1.4;

ICAM-1: 1.3; CX3CL1: 0.9), indicating that CX3CL1 is necessary but not sufficient to

induce transmigration.

5 Results

52

!"#$%&' ()$*+,-.,/0 ()$*+!,12+0 ()$*+3,12+04

0

5

.

6

7

8)$9'($':

!;<!=>=?<;"

@

?9()"A*B9($*#)=!):'C

Figure 12: Neutralizing of CX3CL1, ICAM-1 and VCAM-1 prevents PBMCtransmigration. HUVEC were left unstimulated or were stimulated with IFNγ andTNFα (both 10 ng/ml) for 16 hours and subsequently treated with neutralizing anti-bodies against CX3CL1, ICAM-1 and VCAM-1 or isotype control for 1 hour. Afterwashing, 2x105 PBMC were seeded onto the cell layer for 2 hours at 37°C. Subse-quently, transmigrated cells were counted. Data are shown as mean ± SD of threeindependent experiments. *, p<0.05 versus the untreated control.

It is known that chemokines are internalized upon binding to their receptor. This

process is mediated by β-arrestin and involves the formation of clathrin-coated vesicles

(reviewed in [Moore et al., 2007]. Dynamins are essential for their formation since they

are needed in the transition from a fully formed pit to a pinched-off vesicle. Dynasore is

a small molecule GTPase inhibitor that targets dynamins and therefore blocks dynamin-

dependent endocytosis [Macia et al., 2006]. To address the question, whether

internalization of the ligand-receptor complex also plays a critical role in the

extravasation process, transmigration of PBMC after pretreatment with dynasore was

analyzed. When PBMC were pretreated with dynasore, they did not transmigrate through

a wt-ECV304 cell layer towards soluble CX3CL1 chemokine domain or due to the

interaction with transmembrane CX3CL1 on ECV304 cells (TI = 1.06, TI = 0.95, control

= 1.01) (Figure 13).

5 Results

53

!"!# $%!&'"(#)*)

)*+

,*)

,*+

-*)

-*+ ./01234)526427,01234)58

9(&!':;<(&/;"!=>!$#?

Figure 13: Dynasore prevents transmigration of PBMC. PBMC were pretreatedfor 30 minutes with 100 µM dynasore or left untreated. After washing, cells wereassayed for transmigration through wt- or CX3CL1- ECV304 cell layers cultured intranswell inserts. 2x105 PBMC were seeded onto the confluent cell layer for 2 hoursat 37°C. Subsequently, transmigrated cells were counted. Data are shown as mean ±SD of three independent experiments. *, p<0.05 versus the untransfected control.

The data demonstrate that transmembrane CX3CL1 that is expressed on the cell

surface of endothelial cells not only functions as an adhesion molecule for CX3CR1-

expressing blood mononuclear cells, but can also mediate transmigration through an

endothelial cell layer.

5 Results

54

5.2 Model system for CX3CR1-CX3CL1 function

In order to establish a model system for a more detailed molecular analysis of the

CX3CR1-CX3CL1 system in adhesion and transmigration, two different cell lines were

used that were transfected to express CX3CR1. HEK293 cells were originally derived

from human embryonic kidney cells grown in tissue culture and have an epithelial

morphology. They were already used in a vast variety of receptor studies. Since HEK293

cells did not migrate in our setting, the murine pre-B-cell line L1.2 (also named 300-19)

was used for chemotaxis experiments. Both cell lines do not express CX3CR1

endogenously.

Initial chemotaxis experiments confirmed that CX3CL1 promotes migration of

CX3CR1-transfected L1.2 cells towards soluble CX3CL1 chemokine domain as was

already shown for PBMC (Figure 14). The migration efficiency was comparable to that

obtained for PBMC (Migration Index (MI) = 2.2), with a 2.7 fold increase of migrated

cells. Moreover, HEK293 cells were transfected to express CX3CR1 and assayed for

their ability to adhere to a CX3CL1-expressing ECV304 cell layer, to show that

CX3CR1 mediated adhesion to CX3CL1 under static conditions (Adhesion Index (AI) =

7.4) (Figure 14).

Next, CX3CR1-expressing L1.2 cells were assayed for transmigration through a wt- or

CX3CL1-ECV304 cell layer in the absence or presence of soluble CX3CL1 in the lower

compartment of the transwell chamber. Both soluble and transmembrane CX3CL1

induced transmigration of CX3CR1-L1.2 cells (TI = 1.7 and 1.9, respectively), but not of

wt-L1.2 cells (Figure 15). Simultaneous presence of transmembrane and soluble

CX3CL1 did not further enhance transmigration (TI = 1.9), as was previously seen for

PBMC.

5 Results

55

wt-L1.2 CX3CR1-L1.20

1

2

3 none10nM CX3CL1

*

Mig

ratio

n In

dex

wt-HEK293 CX3CR1-HEK293012345678 wt-ECV304

CX3CL1-ECV304

*

Adhe

sion

Inde

x

Figure 14: L1.2 cells migrate towards CX3CL1 and HEK293 cells adhere toCX3CL1. (left) L1.2 cells were seeded onto a polycarbonate membrane with 8 µmpores in a modified Boyden chamber setting. 10 nM soluble fractalkine chemokinedomain were used to excite chemotaxis for 2 hours at 37°C. Subsequently, migratedcells were counted. (right) Fluorescently labeled wild type or CX3CR1-expressingHEK293 cells were seeded onto wild type ECV304 cells or ECV304 cells express-ing transmembrane CX3CL1, and the fluorescence signal of the adherent HEK293cells was determined after washing. Data are shown as mean ± SD of three inde-pendent experiments. *, p<0.05 versus the controls (random migration of untrans-fected cells and adhesion of untransfected cells, respectively).

!"!

!"#

$"!

$"#

%"!

%"#

&'()*+,!-*.,*/$()*+,!-

0*.,*/$

&'(/$"% *.,*1$(/$"%

( 2 ( 2

33

3

4567089:56'9;7<=7>?@

Figure 15: L1.2 cells transmigrate through an ECV304 cell layer. Wt- andCX3CR1-L1.2 were investigated for transmigration across wt- or CX3CL1-ECV304 cells cultured in trans-well inserts. After washing, 2x105 L1.2 cells wereseeded onto the confluent cell layer for 2 hours at 37°C in the absence or presenceof additional 10 nM CX3CL1 chemokine domain in the lower compartment. Subse-quently, transmigrated cells were counted. Data are shown as mean ± SD of threeindependent experiments. *, p<0.05 versus the random transmigration of untrans-fected cells without addition of stimulus control.

5 Results

56

These studies are well in line with the observations made with primary PBMC and

suggest that the transfected cell lines may be used to investigate chemokine-mediated

cell recruitment. In the next step, the transmigration process was further analyzed with

regard to CX3CL1 expression, signaling and internalization.

I then used HUVEC as a cell layer for transmigration of L1.2 cells. When these cells

were stimulated with IFNγ and TNFα to express endogenous CX3CL1, CX3CR1-L1.2

cells but not wt-L1.2 cells transmigrated through the HUVEC layer (TI = 3.1) (Figure

16).

!"#$%&' ()*(+%#$%&',

%

'

*

-./.0

123!454632"

678.9:;<78";/.41.=0>

?

Figure 16: L1.2 cells transmigrate through a HUVEC cell layer. HUVEC werestimulated with IFNγ and TNFα (both 10 ng/ml) for 16 hours or left untreated. Sub-sequently, wt- and CX3CR1-L1.2 cells were assayed for transmigration across theHUVEC layer. After washing, 2x105 L1.2 cells were seeded onto the confluent celllayer for 2 hours at 37°C. Subsequently, transmigrated cells were counted. Data areshown as mean ± SD of three independent experiments. *, p<0.05 versus the unsti-mulated, untransfected control.

In the previously shown transmigration assay, HUVEC were stimulated to express

CX3CL1 and washed before the assay was performed. To find out whether sufficient

amounts of soluble CX3CL1 to induce transmigration were produced during the assay,

the medium in both compartments of the transwell system was tested in an ELISA for

soluble CX3CL1. As shown in Figure 17, after incubation with pro-inflammatory

5 Results

57

cytokines over night soluble CX3CL1 was produced by HUVEC in an amount that was

shown to promote chemotaxis of PBMC and CX3CR1-L1.2 cells (Figures 9 and 14). In

contrast, 2 hours were not sufficient to produce enough soluble CX3CL1 to mediate

efficient chemotaxis, as there were no detectable differences between stimulated and

unstimulated cells.

!"# $"!!"% !"# $"!!"%&'&

&'(

)'&

)'(

*'&

*'(

+"+,

-./!0102/."

)30405!6%789!6"+ *0409:!,;0<9546+=

>

>

+=?%805"87$8,0@AB@C)

Figure 17: HUVEC produce CX3CL1 after stimulation with cytokines. HU-VEC were stimulated with IFNγ and TNFα (both 10 ng/ml) for 16 hours or left un-treated. Medium was collected, cells were washed three times and new mediumwas given. Cells were left untreated for 2 hours at 37°C and afterwards mediumwas collected and subjected to an ELISA for CX3CL1. Data are shown as mean ±SD of three independent experiments. *, p<0.05 versus the unstimulated control.

Since CX3CL1 is expressed as transmembrane as well as soluble form by endothelial

cells, it is not clear, which form is most important for the transmigration process. The

previous experiment provides an indication that transmembrane CX3CL1 is sufficient to

promote transmigration of CX3CR1-expressing cells. To elucidate the role of

transmembrane CX3CL1, CX3CL1-ECV304 cells were pretreated with a neutralizing

antibody against CX3CL1 for 15 minutes prior to the transmigration assay. Pretreatment

of CX3CL1-ECV304 cells with a neutralizing antibody against CX3CL1 that was added

to the upper compartment of the transwell system (apical site of ECV304 cells) clearly

reduced cell transmigration (TI = 3.3 to TI = 1.5), whereas addition of the antibody to the

5 Results

58

lower compartment (basolateral site) did not suppress cell transmigration (TI = 3.1 to TI

= 2.4) (Figure 18). Simultaneous application of the neutralizing antibody to the upper

and lower compartment did not further decrease transmigration (TI = 3.2 to TI = 1.8).

These data indicate that transmembrane CX3CL1 is crucial for CX3CL1-dependent

transmigration.

none top bottom top+bottom0

1

2

3

4

none anti-CX3CL1 IgG1

* *

Tran

smig

ratio

n In

dex

Figure 18: CX3CL1 inhibition influences transmigration. The neutralizing mon-oclonal antibody to CX3CL1 or isotype control (10 µg/ml) was added to the loweror the upper compartment of transwell inserts containing wt- or CX3CL1-ECV304cells. After 15 minutes, the antibody was removed and the cells were investigatedfor transmigration of CX3CR1-L1.2 cells. Data are shown as mean ± SD of threeindependent experiments. *, p<0.05 versus the untreated, transfected control.

It has been previously demonstrated that pretreatment with soluble CX3CL1 blocks

chemotaxis as well as cell adhesion via CX3CL1, whereas pretreatment with the Gi-

protein inhibitor pertussis toxin (PTX) selectively blocks chemotaxis, but not adhesion

[Fong et al., 1998; Lucas et al., 2003]. Pertussis toxin is an exotoxin with six subunits

released by Bordetella pertussis. Once active, it catalyzes the ADP-ribosylation of the α-

subunit of the Gi-protein and therefore prevents the G-proteins from interacting with G-

protein coupled receptors on the cell membrane. To characterize CX3CL1-mediated

transmigration, CX3CR1-L1.2 cells were pretreated with soluble CX3CL1 or with PTX,

before transmigration through a ECV304 cell layer was assayed. In the transmigration

5 Results

59

assays using CX3CL1-ECV304 cells as substrate, pretreatment with soluble CX3CL1 or

with PTX efficiently suppressed transmigration of CX3CR1-L1.2 cells (TI = 0.85 and

1.14, respectively; control = 2.2) (Figure 19).

!"!# $%& '(&)(*+ !"!# $%& '(&)(*+,

+

-

)

./01(2),3(&)(*+01(2),3

./0*+4- (&)(5+0*+4-

6

%78!'9:;78/:"!<=!>#?

Figure 19: PTX treatment influences transmigration. Wt- and CX3CR1-L1.2cells were pretreated for 30 minutes with soluble CX3CL1 (10 nM) or 2 hours withPTX (100 ng/ml) and subsequently assayed for transmigration across wt- orCX3CL1-ECV304 cells. After washing, 2x105 L1.2 cells were seeded onto the con-fluent cell layer for 2 hours at 37°C. Subsequently, transmigrated cells were count-ed. Data are shown as mean ± SD of three independent experiments. *, p<0.05 ver-sus the transfected, untreated control.

To address the question, whether internalization of the ligand-receptor complex also

plays a critical role in the extravasation process of L1.2 cells, chemotaxis of L1.2 cells

after pretreatment with dynasore was analyzed. When L1.2 cells were pretreated with

dynasore they did not migrate towards soluble CX3CL1 chemokine domain (MI = 1.2,

control = 2.8) (Figure 20).

Taken together, the data indicate that HEK293 cells as well as L1.2 cells transfected

with CX3CR1 are an appropriate model system for the analysis of molecular

determinants of CX3CR1 function with regard to CX3CR1-CX3CL1 interaction.

5 Results

60

Furthermore, transmembrane fractalkine seems to be crucial for CX3CL1-mediated

transmigration, whereas pretreatment of CX3CR1 expressing cells with soluble CX3CL1

inhibits transmigration. Both, the inhibition of G-protein/GPCR interaction and

inhibition of clathrin-coated vesicle formation led to a dramatically reduced chemotaxis

of CX3CR1-L1.2 cells towards CX3CL1. This data indicate that signaling as well as

internalization are essential for transmigration.

!"#$%&' ()*(+%#$%&',

%

'

*

-./.012.34/50

6

78953"8/.:;.10<

Figure 20: Pretreatment with dynasore prevents L1.2 cell chemotaxis. Wt- andCX3CR1-L1.2 cells were pretreated for 30 minutes with 100 µM dynasore or leftuntreated. After washing, cells were assayed for chemotaxis towards 100 ng/mlCX3CL1 chemokine domain in a modified Boyden chamber setting for 2 hours at37°C. Subsequently, migrated cells were counted. Data are shown as mean ± SD ofthree independent experiments. *, p<0.05 versus the untreated, transfected control.

5 Results

61

5.3 Molecular analysis of CX3CR1-CX3CL1 function

To analyze the molecular prerequisites for CX3CL1-mediated adhesion and trans-

migration, the subsequent experiments concentrated on the receptor and addressed

conserved structures in the GPCR superfamily. Analysis showed that there are several

highly conserved motifs known to be important for GPCR signaling (Figure 21). The

aspartate-arginine-tyrosine (DRY) sequence in the second intracellular loop is required

for activation of Gi-proteins, whereas the NPX2-3Y motif located in the seventh

transmembrane region of most GPCR contributes to ligand binding, activation and

internalization of the receptor.

Rhodopsin ALWSLVVLAI ERY VVVCKPMSNF ... AFFAKSAAIY NPVIY IMMNKQFRNCCX3CR1 SIFFITVISI DRY LAIVLAANSM ... ETVAFSHCCL NPLIY AFAGEKFRRYCXCR1 GILLLACISV DRY LAIVHATRTL ... EILGILHSCL NPLIY AFIGQKFRHGCCR5 GIFFIILLTI DRY LAVVHAVFAL ... ETLGMTHCCI NPIIY AFVGEKFRNYGnRHR PAFMMVVISL DRS LAITRPLALK ... FLFAFLNPCF DPLIY GYFSL CHRM1 SVMNLLLISF DRY FSVTRPLSYR ADRB1 SIETLCVIAL DRY LAITSPFRYQ ... NWLGYANSAF NPIIY CRSPDFRKAFCXCR6 SMLILTCITV DRF IVVVKATKAY ... EAIAYLRACL NPVLY AFVSLKFRKN

Figure 21: Alignment of conserved motifs in GPCR superfamily. Differentmembers of class A or rhodopsin-like GPCR were aligned to show two importantconserved motifs in the GPCR superfamily. The DRY motif is located in the sec-ond intracellular loop and always shows the characteristic arginine-residue, where-as the aspartate-residue is sometimes changed to be glutamic acid. The tyrosineresidue is unfrequently altered to either serine or phenylalanine. The NPX2-3Y mo-tif is located in the seventh transmembrane domain. The proline and the tyrosineresidues are seldom found to be changed, whereas the asparagine residue some-times is altered to aspartate. The two or three unspecified amino acids are in mostcases leucine, isoleucine or valine. Abbreviations and Uniprot.org Accession num-bers: Rhodopsin: P30968, CX3CR1: P49238, CXCR1: P25024, CCR5: P51681,GnRHR: Gonadotropin-releasing hormone receptor, P30968, CHRM1: Muscarinicacetylcholine receptor M1, P11229, ADRB1: Beta-1 adrenergic receptor, P08588,CXCR6: O00574.

Additionally, GPCR typically carry several serine residues within the intracellular C-

terminal region, which can become phosphorylated by G-protein-coupled receptor

5 Results

62

kinases and mediate interaction with β-arrestins, internalization via clathrin-coated

vesicles and subsequent desensitization of the receptor towards its ligand. The typical

DRY sequence, the NPX2-3Y motif and several C-terminal Ser-residues are also found

within CX3CR1 and very likely play a role in CX3CR1 signaling and function. To

analyze CX3CR1 function on molecular basis, CX3CR1 variants with altered conserved

motifs were constructed. I changed the DRY motif into DNY (R127N), the NPX2-3Y

motif into APX2-3Y (N289A) and NPX2-3A (Y293A), and truncated the intracellular

stalk at S319 (S319X) to remove all intracellular serine-residues. In Figure 22 all

locations for changes are shown.

Figure 22: Overview of introduced changes in CX3CR1. CX3CR1 carries anumber of conserved sequences, which can be found throughout GPCR superfami-ly, including a DRY motif in the second intracellular loop, a NPX2-3Y motif in theseventh transmembrane domain and several serine residues at the C-terminus. Toalter these motifs, CX3CR1 was mutated at the indicated sites. R127 in the secondintracellular loop was mutated to asparagine (R127N), N289 and Y293 in the sev-enth transmembrane domain were changed into alanine (N289A and Y293A, re-spectively), and the intracellular C-terminus was truncated before S319 (S319X).Locations for changes are indicated with filled circles.

5 Results

63

The cDNA for CX3CR1 variants was constructed using two-fragment ligation (for

R127N, N289A and Y293A) or a simple PCR strategy (for S319X) and cloned into

pcDNA3.1+ expression vector. All constructs were successfully cloned and confirmed by

sequencing. To determine whether the CX3CR1 variants could be expressed, HEK293

cells were transiently transfected and tested for expression of CX3CR1 using flow

cytometry. As shown in Figure 23, all constructed variants could be readily detected on

the cell surface of HEK293 cells 2 days after transfection, all expressed with similar

intensity. To examine whether the transfected receptor variants bind CX3CL1, cells were

incubated with CX3CL1 that was fused to the Fc-part of human IgG1. This construct was

then detected by flow cytometry using an anti-Fc-antibody (Figure 23). Ligand binding

was still detectable for the R127N and S319X variants, but absent in the N289A and

Y293A variants. Since ligand binding is the crucial step for mediation of function, I

decided to concentrate further functional characterization on the R127N and S319X

mutations.

Expression Binding0

1

2

3mockCX3CR1N289AY293AR127N

* * S319X

Log

of M

ean

Fluo

resc

ence

Inte

nsity

Figure 23: Expression and binding of CX3CR1 variants. Expression of the re-ceptor variants was controlled by flow cytometry using PE-conjugated rat anti-CX3CR1 antibody. Ligand binding was analyzed by flow cytometry using a recom-binant CX3CL1-Fc construct as a ligand. Receptor expression and ligand bindingwere measured as the mean fluorescence intensity increase compared to control.After logarithmic transformation, the data were summarized as means plus SD fromfive independent experiments. Expression in mock cells was used as a covariate. *,p<0.05 versus CX3CR1 transfected control.

5 Results

64

The R127N and S319X receptor variants were transfected in HEK293 cells using

lipofectamine. After 48 hours, cells were selected for receptor expression by G418. Only

cells that express the plasmid have the resistance for this antibiotic. After several days of

G418 preselection, mixed clones were diluted and seeded on a 96-well-plate at the

concentration of 1 cell/well. Single clones were grown and subsequently subcultivated.

Receptor expression was controlled using flow cytometry. Clones matching the

expression of CX3CR1-transfected HEK293 cells were chosen for further experiments

(Figure 24).

Rela

tive C

ell

Num

ber


Isotype

CX3CR1

R127N

S319X

Figure 24: Stably transfected HEK293 cells. CX3CR1 and its R127N and S319Xvariants were stably expressed in HEK293 cells and selected for comparable sur-face expression by flow cytometry using PE-conjugated rat anti-CX3CR1. Duringall further experiments, cells were constantly controlled for stable receptor expres-sion. Data are shown as a representative histogram.

Following ligand binding, most G-protein coupled receptors rapidly become

internalized which allows to control desensitization towards the ligand [Ben-Baruch et

al., 1995; Kraft et al., 2001]. To further examine the influence of conserved motifs on

receptor function, fluorescently labeled CX3CL1 chemokine domain that allows the

visualization of internalization processes was used. HEK293 cells expressing the

indicated receptor variants were incubated with 10 nM of AlexaFluor647-labeled

CX3CL1 chemokine domain for 5 - 30 minutes at 37°C. Immediately after stimulation,

5 Results

65

cells were kept on ice to avoid further internalization. As control that the fluorescence

increase was due to ligand uptake rather than ligand binding, aliquots of cells were

incubated under the same conditions, but in the continuous presence of NaN3 during the

whole experiment in order to prevent ligand uptake.

Rela

tive C

ell

Num

ber


none

AF-CX3CL1 + NaN3

AF-CX3CL1

0 # 10 1# 20 2# 30

0.0

2.#

#.0

(.# wt CX3CR1

R12(. /310X

*

*

2i4e 64in8

9:;

ore

>?en?e In?re

a>e 6

BC

8

*

!

!

Figure 25: CX3CR1 internalizes upon CX3CL1 treatment and mediatesCX3CL1 uptake. (top) HEK293 cells expressing the different receptor variantswere incubated with AlexaFluor647-labeled CX3CL1 (AF-CX3CL1) at 37°C in theabsence or presence of NaN3 (0.2%). Subsequently, cells were analyzed for bindingand uptake of CX3CL1 by flow cytometric measurement of fluorescence. A repre-sentative histogram of the fluorescence signal for CX3CR1-HEK293 cells isshown. (bottom) HEK293 cells expressing the different receptor variants were incu-bated with AF-CX3CL1 at 37°C for the time periods indicated. The fluorescenceintensity for the different HEK293 mutants was expressed in relation to that of thecontrol receiving no AF-CX3CL1, and is presented as mean ± SD from three inde-pendent experiments. *, p<0.05 versus the untransfected control. ✝, p<0.05 versusCX3CR1-HEK293 cells.

5 Results

66

As shown in Figure 25, expression of CX3CR1 or the R127N mutation led to a time-

dependent increase in fluorescence, as measured by flow cytometry. This increase was

almost completely abolished when cells were continuously treated with NaN3. No

specific fluorescence uptake was seen with wt-HEK293 cells, and increase in

fluorescence was clearly reduced when the S319X variant was expressed, indicating that

the receptor’s C-terminus contributes to receptor-mediated ligand uptake.

Since the concentration of intracellular free calcium rapidly increases upon activation

of most chemokine receptors including CX3CR1 [Imai et al., 1997], I decided to test

whether the receptor variants were able to mediate intracellular calcium signals.

Therefore, HEK293 cells expressing the indicated receptor variants were loaded with the

fluorescent calcium indicator Fluo3-AM. While Fluo3-AM, in contrast to Fluo3, can

traverse the cell membrane, the molecule itself does not bind Ca2+. Once the dye is inside

the cell, it is hydrolyzed to Fluo3 by endogenous esterases. When bound to calcium

Fluo3 absorbs at 506 nm and emits at 526 nm [Kao et al., 1989; Minta et al., 1989]. This

signal can be measured in a fluorescence reader to visualize the intracellular calcium

flux.

On stimulation with 10 nM CX3CL1 chemokine domain, HEK293 cells expressing

CX3CR1 showed a rapid increase of the fluorescence signal of about 0.25 mM Ca2+,

peaking at about 5 seconds after stimulation and lasting for about 30 seconds. Wt-

HEK293 cells or HEK293 cells expressing the R127N variant, however, did not show

any increase in fluorescence signal intensity, whereas the fluorescence signal of HEK293

cells expressing the S319X was slightly lower (about 0.2 mM Ca2+) than that of the

wildtype receptor (Figure 26) after treatment with soluble CX3CL1. These data indicate

that the induction of intracellular calcium transients by CX3CR1 requires the integrity of

the receptor’s DRY motif, but is independent of its C-terminus.

5 Results

67

0 50 100 150 200 250

wt CX3CR1

Time [s]

R127N

S319X 0.5

mM

Ca

2+

Figure 26: CX3CR1 and S319X, but not R127N mediate intracellular calciumflux. HEK293 cells expressing the indicated receptor variants were loaded with cal-cium indicator Fluo3-AM and changes in fluorescence intensity upon treatmentwith soluble CX3CL1 chemokine domain (10 nM at 100 seconds) were recorded.Loading was controlled using digitonin for maximal and EGTA for minimal signal.Length of arrow indicates an increase of 0.5 mM Ca2+. Results shown are represen-tative for three independent experiments.

To explore the functional relevance of the R127N and S319X variants of CX3CR1 for

cell recruitment, stably transfected HEK293 cells were tested for their ability to adhere to

CX3CL1-expressing ECV304 cells and HUVEC under static and flow conditions. Wt-

ECV304 or CX3CL1-ECV304 cells were grown to confluency, and fluorescently labeled

HEK293 cells expressing the indicated receptor variants were either seeded onto the

layer for static adhesion or directed over the cell layer with a constant shear stress rate of

2 dyne/cm2 for adhesion under flow conditions. As shown in Figure 27, disruption of

conserved motifs in CX3CR1 did neither change its capability to mediate adhesion under

static conditions (AI = 7.35; 6.97 and 7.80, respectively) nor under flow conditions (AI =

13.34; 15.49 and 13.12, respectively).

5 Results

68

wt CX3CR1 R127N S319X 0.0

2.5

5.0

7.5

10.0

wt-ECV304 CX3CL1-ECV304

HEK293

Adhesio

n Index

* *

*

wt CX3CR1 R127N S319X 0

5

10

15

20

*

*

*

HEK293

Adhesio

n Index

Figure 27: Receptor variants mediate adhesion under static and flow condi-tions. The ability of receptor variants to mediate adhesion to CX3CL1 was testedunder static (top) and flow (bottom) conditions. (top) For static adhesion assays,calcein-labeled HEK293 cells expressing the indicated receptor variants were seed-ed onto wt- or CX3CL1-ECV304 cells and after washing, the fluorescence signal ofthe adhering cells was measured. (bottom) For flow adhesion experiments, labeledHEK293 cells expressing the indicated receptor variants were perfused over a layerof cytokine-stimulated HUVEC for 1 minute and adherent cells were counted. Dataare shown as mean ± SD of three independent experiments. *, p<0.05 versus theuntransfected control.

Since calcium signals are required for proper migration, I expected that cells

expressing the R127N variant of CX3CR1 would not migrate towards CX3CL1. As

HEK293 cells did not show migration properties, it was decided to use the murine pre-B-

cell line L1.2 that was stably transfected with the indicated receptor variants. L1.2 cells

do not express endogenous CX3CR1, but are able to mediate migration towards

chemokines when transfected with chemokine receptors, as was previously indicated in

5 Results

69

section 5.2. L1.2 cells were transfected using the AMAXA system and selected with

G418. L1.2 cell clones expressing the receptor variants were chosen according to their

expression level similar to that of L1.2 cells stably expressing CX3CR1.

Chemotaxis experiments were performed to investigate, whether the R127N and

S319X variants of CX3CR1 would allow migration in response to soluble CX3CL1. In a

modified Boyden chamber chemotaxis assay, only the CX3CR1-expressing cells (MI at

10 nM = 2.46), but not the cells with altered receptors migrated towards increasing

concentrations of CX3CL1 (Figure 28) (MI at 10 nM CX3CL1: R127N-CX3CR1 = 1.07

and S319X-CX3CR1 = 1.02). Again, 10 nM CX3CL1 was the optimal concentration for

induction of chemotaxis in CX3CR1-L1.2 cells.

I then examined the ability of CX3CR1 variants to induce transmigration through the

ECV304 cell layer. When CX3CR1-L1.2 cells were seeded onto CX3CL1-ECV304

cells, transmigration was increased (TI = 2.51). By contrast, transmigration was even

below baseline when L1.2 cells expressed the R127N or S319X-mutant (TI = 0.28 and

0.25, respectively) (Figure 28).

To confirm these data in a more physiologic setting, cell layers of cytokine-stimulated

HUVEC instead of ECV304 cells were used for transmigration assays. As seen in Figure

29, CX3CR1 expression increased transmigration of L1.2 cells (TI = 3.12), whereas the

expression of R127N-CX3CR1 suppressed transmigration (TI = 0.61) (S319X-CX3CR1

was not investigated). These results suggest that receptor function via the DRY motif and

the C-terminus are required for transmigration. In the absence of these motifs, however,

transmigration is blocked, which could be explained by the retention of leukocytes

adhering to transmembrane CX3CL1.

5 Results

70

!

"

#

$

!%!" !%" " "! "!!

&'()"%# *+$*,"()"%# ,"#-.()"%# /$"0+()"%#

1

11

1

!*+$*)"2*34564'78'9342:4;<

;9=78'9342>4?6@

!t CX3C&1 &12)* +31,X0

1

2

3!t-/C0304CX3CL1-/C0304

*

* *

L142

Tran

smig

ratio

n In

dex

Figure 28: Receptor variants do not mediate (trans-)migration. The ability ofreceptor variants to mediate migration (top) and transmigration (bottom) towardsCX3CL1 was tested. (top) L1.2 cells expressing the indicated receptor variantswere tested for their chemotactic response towards increasing concentrations of sol-uble CX3CL1 (0.1 nM - 100 nM) in a modified Boyden chamber assay. (bottom)L1.2 cells expressing the indicated receptor variants were studied for transmigrationthrough CX3CL1- ECV304 cell layers. Data are shown as mean ± SD of three inde-pendent experiments. *, p<0.05 versus the untransfected controls (unstimulated oruntransfected and unstimulated).

5 Results

71

wt CX3CR1 R127N0

1

2

3

4

L1.2

Tran

smig

ratio

n In

dex

*

Figure 29: Receptor variant R127N does not mediate transmigration throughHUVEC. HUVEC were stimulated with IFNγ and TNFα (both 10 ng/ml) for 16hours and subsequently probed for transmigration of L1.2 cells expressing the indi-cated receptor variants. Data are shown as mean ± SD of three independent experi-ments. *, p<0.05 versus the untransfected control.

The extravasation process requires the cell to find an appropriate site for transcellular

diapedesis. When L1.2 cells expressing the indicated variants were co-incubated with wt-

or CX3CL1-ECV304, L1.2 cells expressing the wildtype receptor crawled on the cell

layer expressing CX3CL1 and extended pseudopods (69% of cells). Formation of

pseudopods might help the cells to find a site for transmigration and was not present

when the ligand was not expressed on the cell layer (13.52 to 14.51% of cells). When

L1.2 cells expressing the R127N or S319X-CX3CR1 variants were assayed, the

pseudopod formation rate was reduced (24.15% and 12.76% of cells, respectively)

(Figure 30).

5 Results

72

!" #$%#&' &'()* +%',$-'-(-%-.-/-0-)-1-,-

!"23#4%-.#$%#5'23#4%-.

6

5'7(

89:;9<"=>?=8@9AB>C>B=D>:EF"G><

Figure 30: Receptor variants do not mediate pseudopod formation. L1.2 cellsexpressing the indicated variants were co-incubated with a wt- or CX3CL1-ECV304 cell layer. For 5 minutes every 10 seconds a picture of a randomly chosenfield was taken. 100 cells per field were analyzed for pseudopod formation. In thelower panel a representative formation of a pseudopod of a CX3CR1-L1.2 cell on aCX3CL1-ECV304 cell layer is shown (white scale bars represent 3 µm). Data areshown as mean ± SD of three independent experiments. The asterisk indicates sta-tistically significant differences versus the untransfected control, p<0.05.

5 Results

73

5.4 Involvement of ADAMs in CX3CR1-CX3CL1 function

I then questioned whether the activity of CX3CL1 sheddases on ECV304 cells is

required for transmigration towards CX3CL1. To address this issue, the ECV304 cell

layer was pretreated with GW280264X, an inhibitor of the metalloproteinases ADAM10

and ADAM17 that shed transmembrane CX3CL1 [Garton et al., 2001; Hundhausen et

al., 2003]. To selectively address the role of the proteases on the ECV304 cell layer, the

inhibitor was removed prior to starting the transmigration assay with the addition of L1.2

cells. As shown in Figure 31, the inhibitor did not affect L1.2 cell recruitment via soluble

CX3CL1 chemokine domain (TI = 1.78; DMSO control: TI = 1.82), but clearly blocked

transmigration in response to transmembrane CX3CL1 (TI = 0.97; DMSO control: TI =

1.66). Moreover, this inhibition of transmembrane CX3CL1 activity could not be

overcome by addition of soluble CX3CL1 chemokine domain as chemoattractant (TI =

1.21; control: TI = 1.66). Residual transmigration in the absence of CX3CL1 was not

affected by GW280264X, indicating that there was no general effect on the integrity of

the cell layer. These results indicate that the activity of ADAMs is only required for

transmigration in response to transmembrane CX3CL1, but not to soluble CX3CL1.

The following set of experiments was performed to confirm the effect of the inhibitor

with PBMC. CX3CL1-ECV304 cells were coincubated in the absence or presence of

PBMC to investigate whether PBMC would affect CX3CL1 surface expression on the

ECV304 cells. Therefore, PBMC were seeded onto a ECV304 cell layer for 3 hours at

37°C and were then tested for their CX3CR1 surface expression in flow cytometry. To

differentiate between PBMC and ECV304 cells, cells were gated for size. The surface

expression level was reduced in the presence of PBMC (relative fluorescence intensity =

0.69) (Figure 32). Pretreatment of CX3CL1-ECV304 cells with GW280264X, however,

considerably increased CX3CL1 surface expression (relative fluorescence intensity =

3.26), which was not affected when the CX3CL1-ECV304 cell layer was exposed to

PBMC (relative fluorescence intensity = 3.40).

5 Results

74

!"!# $%&'()'*+ !"!# $%&'()'*+,-,

,-.

+-,

+-.

/-,

/-. 0123 45/6,7/8(

99

:;<='>),8 '()'*+<='>),8

?@A!&BCD@A;C"!%E!F#G

Figure 31: Pharmacologic inhibition of ADAM10/17 abrogates transmem-brane CX3CL1-dependent L1.2 cell transmigration. Wt- or CX3CL1-ECV304cells were grown on transwell inserts and pretreated with GW280264X (10 µM) orDMSO control for 1 hour. After removal of the inhibitor, the lower compartmentsof the transwell systems received soluble CX3CL1 chemokine domain (10 nM) orwere left without stimulus, and subsequently CX3CR1-L1.2 cells were assayed fortransmigration. Data are shown as mean ± SD of three independent experiments. *,p<0.05 versus the mock-treated control.

DMSO GW280624X 0

1

2

3

4 none

PBMC

fold

exp

ressio

n o

f co

ntr

ol

*

Figure 32: CX3CL1 surface expression is reduced when incubated withPBMC. CX3CL1-ECV304 cells were pretreated with GW280264X (10 µM) orDMSO for 1 hour, washed and then co-incubated with freshly prepared PBMC for3 hours. Cells were harvested and analyzed for CX3CL1 surface expression by flowcytometry. The mean fluorescence intensity was expressed in relation to that ofCX3CL1-ECV304 cells receiving no inhibitor and no PBMC and data are shown asmean ± SD of three independent experiments. *, p<0.05 versus the control.

5 Results

75

!t-$C&3() CX3CL1-$C&3()(1-3)./012

345678-1(/-)X

9

A;<e

sion

In;e

D

wt#$C&30) CX3C+1#$C&30)0.0

0.5

1.0

1.5

2.0

2.5 01S34526072)X

*

Tran

smi@

ratio

n In

DeF

Figure 33: PBMC adhere, but do not transmigrate towards CX3CL1 whenADAM10/17 are inhibited. (top) Wt- or CX3CL1-ECV304 cells were pretreatedwith DMSO as control or GW280264X (10 µM) for 1 hour, washed and subse-quently analyzed for adhesion of PBMC under flow. (bottom) Wt- or CX3CL1-ECV304 cells were grown on transwell inserts, pretreated with DMSO as control orGW280264X (10 µM) for 1 hour and subsequently probed for transmigration ofPBMC. Data are shown as mean ± SD of three independent experiments. *, p<0.05versus the mock-treated controls.

To examine whether the increased CX3CL1 surface expression after treatment with

the inhibitor would affect cell recruitment, adhesion assays were performed. CX3CL1-

mediated adhesion to ECV304 cells was considerably increased when the cells were

pretreated with the ADAM10/17 inhibitor (AI = 7.0; DMSO control: AI = 4.25) (Figure

33). However, despite increased CX3CL1 surface expression and increased CX3CL1-

mediated adhesion, CX3CL1-mediated transmigration through the ECV304 cell layer

was completely suppressed by the inhibitor (TI = 1.15; DMSO control: TI = 2.05)

5 Results

76

(Figure 33). Residual transmigration in the absence of CX3CL1 was not affected by

GW280264 (TI = 1.15).

As shown in Figure 34, experiments were repeated with HUVEC, demonstrating that

PBMC adhesion to cytokine-stimulated HUVEC is profoundly increased (AI = 3.07;

DMSO control: AI = 2.01), whereas transmigration is completely suppressed by the

inhibitor (TI = 0.94; DMSO control: TI = 2.02).

!"#$%&!'($)*+ ,-.!+/+0.-"1

2

3

4

56789:;3<1=35>

?*@)#%A"+,"*)B

C

!"#$%&!'($)* +,-!./.0-,"121

123

421

423

521

523

6789:;5<1=5>?

0@("#&%A@($%B".+"*)C

D

Figure 34: PBMC adhere, but do not transmigrate towards CX3CL1 whenADAM10/17 are inhibited. (top) HUVEC were stimulated with IFNγ and TNFα(both 10 ng/ml) or left unstimulated for 16 hours, and were pretreated with DMSOas control or GW280264X (10 µM) for 1 hour, washed and subsequently analyzedfor adhesion of PBMC under flow. (bottom) HUVEC were grown on transwell in-serts, stimulated with IFNγ and TNFα (both 10 ng/ml) or left unstimulated for 16hours, and were pretreated with DMSO as control or GW280264X (10 µM) for 1hour and subsequently probed for transmigration of PBMC. Data are shown asmean ± SD of three independent experiments. *, p<0.05 versus the mock-treatedcontrol.

5 Results

77

The importance of ADAM10 and ADAM17 in CX3CL1-dependent transmigration

was further investigated by shRNA-mediated silencing of these proteases. Martin Hess of

our laboratory generated a lentiviral expression vector for shRNA that was used for

efficient and sustained downregulation of ADAM10 and ADAM17 surface expression on

ECV304 cells. By ELISA, he confirmed that silencing of ADAM10 led to a reduced

release of soluble CX3CL1 and accumulation of CX3CL1 in the cell lysates. Silencing of

ADAM17 had no comparable effect on the constitutive release of CX3CL1, which is

consistent with the previous observation that this protease is implicated in the PMA-

stimulated release of CX3CL1, but not in the constitutive shedding of the chemokine.

CX3CR1-mediated transmigration through CX3CL1-expressing ECV304 cell layers

(TI = 1.95) was almost completely suppressed when ADAM10 expression was silenced

(TI = 1.02), and moderately reduced when ADAM17 expression was downregulated (TI

= 1.34) (Figure 35).

!" #$%#&'()*+,-./0 #$%#&'(1'2 #$%#&'(1'3242

245

'42

'45

642

645!"7&'46#$%#8'7&'46

9

:#;%2<

=+,>)-?@+,"?A>(B>C0D

Figure 35: ADAM10, but not ADAM17 silencing leads to an almost completesuppression of CX3CL1-mediated transmigration. Stably transduced CX3CL1-ECV304 cells were investigated for transmigration of wt- and CX3CR1-L1.2 cells.After washing, 2x105 L1.2 cells were seeded onto the confluent ECV304 cell layerfor 2 hours at 37°C. Subsequently, transmigrated cells were counted. A10 indicatesthe transduction of CX3CL1-ECV304 cells with shRNA for ADAM10, and A17 in-dicates the transduction with shRNA for ADAM17. Data are shown as mean ± SDof three independent experiments. *, p<0.05 versus the untransfected control.

5 Results

78

These results indicate that the proteolytic activity of ADAM10, and to a lesser degree

of ADAM17, decreases CX3CL1 surface expression and CX3CL1-mediated adhesion

and is required for transmigration in response to CX3CL1.

5 Results

79

5.5 CXCR6-CXCL16 function

CXCL16 shares a number of structural and functional properties with CX3CL1. It is

expressed as transmembrane variant and shed by ADAM10 resulting in a soluble

CXCL16 variant that mediates chemotaxis of CXCR6 expressing T-cells [Matloubian et

al., 2000; Abel et al., 2004]. When CXCR6 was analyzed for conserved motifs, a DRF

motif instead of a DRY motif was found in the second intracellular loop (Figure 21).

This alteration was already described for the 5-HT2βPan -receptor of Panulirus interruptus

(spiny lobster) and implicated in its constitutive activity [Clark et al., 2004]. This raises

the question which role the DRY(F) motif would play for the function of CXCR6.

To investigate the function of CXCR6, mononuclear cells from peripheral blood of

healthy donors were probed for CXCR6 expression by flow cytometry using a PE-

labeled antibody. CXCR6 was previously described to be expressed on CD4+ T-helper 1,

CD8+ T-cytotoxic and T-regulatory 1 subsets of T-cells, smooth muscle cells, dendritic

cells, B-cells, macrophages and subsets of natural killer cells [Matloubian et al., 2000;

Sharron et al., 2000; Kim et al., 2001; Wilbanks et al., 2001; Hofnagel et al., 2002; Sato

et al., 2005]. Depending on donor and cell type, the expression level highly varied from

nearly absence of detectable CXCR6 expression and high CXCR6 expression. For all

donors the strongest expression of CXCR6 could be seen on T-cell subsets (Figure 36).

A fundamental step in leukocyte recruitment is the firm adhesion on the endothelium

lining the blood vessel. CXCR6, expressed on T-cells, has been reported to be involved

in CXCL16 mediated T-cell recruitment to inflammatory sites [Galkina et al., 2007].

Freshly prepared PBMC from donors expressing high or low levels of CXCR6 were

tested for their adhesion to CXCL16-expressing cell layers under static conditions. When

PBMC were co-incubated with CXCL16-ECV304 cells and subsequently washed, no

adhesion could be observed. This effect was independent from the level of CXCR6

expression (low expression: AI = 0.85, high expression: AI = 0.93, control: AI =

1.0)(Figure 37).

5 Results

80

Figure 36: CXCR6 expression varies based on donor. Freshly isolated bloodmononuclear cells of healthy donors were probed for their expression of CXCR6using a PE-labeled anti-hCXCR6 antibody. The grey shaded curve represents theisotype control, the black unshaded curve represents hCXCR6 expression of notgated PBMC. Donor A expresses CXCR6 on PBMC subsets (left), whereas noCXCR6 was detected on PBMC of donor B (right). Data are shown as a representa-tive histogram.

!"!#$%&' !"!#$()*(+,+

+,-

.,+

.,- '/01!23+4!"!5.$01!23+4

678!

9:(;<)&=>?=:;@

Figure 37: PBMC do not adhere to CXCL16 under static conditions. Freshlyisolated PBMC from donors with a high or low expression of CXCR6 were assayedfor adhesion under static conditions on wt-ECV304 or CXCL16-ECV304 cell layer.PBMC were stained with CalceinAM for 15 minutes. Excess dye was removed bywashing, cells were suspended to a concentration of 5x105 cells/ml and seeded ontowt- or CXCL16-ECV304 cells. After washing, the fluorescence signal of the adher-ing cells was measured. Data are shown as mean ± SD of three independentexperiments.

5 Results

81

Due to the high variability of CXCR6 expression on PBMC, I then tested HEK293

cells stably expressing CXCR6 for their ability to adhere to CXCL16 expressing cells

under static conditions. Wt-ECV304 or CXCL16-ECV304 cells were grown to

confluency, and fluorescently labeled wt-HEK293 and CXCR6-HEK293 were seeded

onto the cell layer. As shown in Figure 38, no adhesion of CXCR6 expressing cells to

CXCL16 expressing cells could be observed under static conditions (AI = 1.7; control:

AI = 1.0).

!"#$%&'() *+*,-#$%&'()./.

./0

1/.

1/0

'/.

'/0!"#%*2).3*+*41-#%*2).3

56789:;<=><68?

Figure 38: CXCR6-expressing cells do not adhere to CXCL16 under staticconditions. Wt-HEK293 and CXCR6-HEK293 cells were assayed for adhesion un-der static conditions on wt-ECV304 or CXCL16-ECV304 cell layer. HEK293 cellswere stained with CalceinAM for 15 minutes. Excess dye was removed by wash-ing, and cells were suspended to a concentration of 5x105 cells/ml and seeded ontowt- or CXCL16-ECV304 cells. After washing, the fluorescence signal of the adher-ing cells was measured. Data are shown as mean ± SD of three independentexperiments.

As CXCL16 is described to bind to CXCR6, I performed a binding assay with 6xHis-

tagged recombinant CXCL16 chemokine domain. Cells were incubated with

CXCL16-6xHis in the absence or presence of a 10 fold excess of CXCL16 chemokine

domain. Bound CXCL16-6xHis was then detected with an anti-6xHis-tag-antibody.

When PBMC of a donor with high expression of CXCR6 were analyzed, binding was

found that could be diminished by competition with untagged CXCL16 chemokine

domain. In contrast, PBMC from donors with a low expression of CXCR6 did not bind

5 Results

82

CXCL16-6xHis (Figure 39).

Figure 39: CXCR6 binds CXCL16-6xHis. CXCL16 binding to PBMC was ana-lyzed by flow cytometry using recombinant 6xHis-tagged CXCL16 chemokine do-main as ligand in the absence or presence of a 10-fold excess of untagged CXCL16chemokine domain. The binding was detected using a murine anti-6xHis-tag-anti-body and an anti-mouse-PE antibody. The pink curve represents the untreated con-trol, the blue curve represents detected binding of 6xHis-tagged CXCL16 onPBMC and the green curve represents 6xHis-tagged CXCL16 binding when un-tagged CXCL16 is present for competition. Donor A expresses CXCR6 on PBMCsubsets (left), whereas no CXCR6 was detected on PBMC of donor B (right).Results shown are representative for three independent experiments.

Albeit its inability to mediate adhesion, there remains the possibility that the binding

of CXCL16 to CXCR6 would mediate chemotaxis. Freshly prepared PBMC were

assayed for chemotaxis in response to soluble CXCL16 chemokine domain (1 - 100 nM)

in a modified 48-well Boyden chamber setting. When PBMC highly expressing CXCR6

were used, increased migration could be observed (MI = 1.85 at 10 nM CXCL16).

Chemotaxis towards CXCL16 chemokine domain was dose-dependent and showed the

characteristic bell shaped curve reported for chemokines peaking at 10 nM soluble

CXCL16 chemokine domain (Figure 40). Based on this curve, all further experiments

were performed with 10 nM soluble CXCL16 chemokine domain. In contrast, when

CXCR6 expression could not be detected on PBMC, no migration towards CXCL16

chemokine domain was observed (MI = 0.98 at 10 nM CXCL16). These data indicate

5 Results

83

that CXCR6-mediated signaling of chemotaxis occurs despite of the alteration of the

conserved DRY motif into DRF.

!"!# $%!&% $'%!&% $''%!&%'('

'()

$('

$()

*(' +

+

,-,.$/

&012340"!%5!6#7

!"!# $%!&% $'%!&% $''%!&%'('

'()

$('

$()

*('

+,+-$.&/0123/"!%4!5#6

Figure 40: PBMC expressing CXCR6 chemotactically migrate towardsCXCL16. PBMC were seeded onto a polycarbonate membrane with 8 µm pores ina modified Boyden Chamber setting. 1-100 nM soluble CXCL16 chemokine do-main were used to excite chemotaxis for 2 hours at 37°C. Subsequently, migratedcells were counted. (left) Donor A expressed CXCR6 on PBMC subsets, whereasno CXCR6 was detected on PBMC of donor B (right). Data are shown as mean ±SD of three independent experiments. *, p<0.05 versus the unstimulated control.

For a better understanding of the DRF motif in CXCR6 function, different receptor

variants were created: the DRF motif was changed into DNF (R127N) consistent with

the CX3CR1 experiments, and the DRY motif (F128Y) was reconstituted. The cDNA for

CXCR6 variants was constructed using two-fragment ligation and cloned into

pcDNA3.1+ expression vector. All constructs were successfully cloned and confirmed by

sequencing.

To determine whether the mutated CXCR6 variants would be expressed, HEK293

cells were transiently transfected and tested for expression of CXCR6 using flow

cytometry. As shown in Figure 41, all constructed variants could be readily detected on

the cell surface of HEK293 cells 2 days after transfection. The expression level of

CXCR6 seemed not to be influenced by the alteration.

5 Results

84

Figure 41: Surface expression and ligand binding of CXCR6 variants. (left)Expression of the receptor variants was controlled by flow cytometry using a PE-conjugated anti-CXCR6 antibody. (right) Ligand binding was analyzed by flow cy-tometry using a recombinant CXCL16-6xHis construct as a ligand. The bindingwas detected using a murine anti-6xHis-tag-antibody and an anti-mouse-PE anti-body. The grey shaded curve represents unstained wildtype HEK293 cells, the pinkcurve represents the stained mock transfected control, the blue curve representsstained CXCR6 transfected cells, the green curve represents stained R127N varianttransfected cells, and the orange curve represents stained F128Y variant transfectedcells. Data are shown as a representative histogram.

Since ligand binding can be affected by mutating CXCR6, transiently CXCR6-

transfected HEK293 cells were tested for binding of a recombinant CXCL16-6xHis

construct. No differences in ligand binding could be observed between the wildtype

receptor and its variants (Figure 41).

To test whether the variants of CXCR6 are still able to mediate intracellular calcium

transients, HEK293 cells expressing the indicated receptor variants were loaded with the

fluorescent calcium indicator Fluo3-AM. On stimulation with 10 nM CXCL16, HEK293

cells expressing CXCR6 showed a calcium signal of about 0.05 mM Ca2+, peaking at

about 10 seconds after stimulation and lasting for about 45 seconds. Wt-HEK293 cells or

HEK293 cells expressing the R127N variant, however, did not show any signal.

Interestingly, the calcium signal of HEK293 cells expressing the F128Y variant was

5 Results

85

slightly decreased (about 0.04 mM Ca2+) compared to the response of cells expressing the

wildtype receptor (Figure 42). These data indicate that the induction of intracellular

calcium transients by CXCR6 requires the arginine residue of the receptor’s DRF motif,

but the phenylalanine residue is less critical.

! "! #!! #"! $!! $"!

%&

'(')*

)#$+,

-#$./

01234567

!8$"49:4';$<

Figure 42: CXCR6 and F128Y, but not R127N mediate intracellular calciumflux. HEK293 cells expressing the indicated receptor variants were loaded with cal-cium indicator Fluo3-AM and changes in fluorescence intensity upon treatmentwith soluble CXCL16 chemokine domain (10 nM at 100 seconds) were recorded.Loading was controlled using digitonin for maximal and EGTA for minimal signal.The length of the arrow indicates an increase of 0.25 mM Ca2+. Results shown arerepresentative for three independent experiments.

To explore the functional relevance of the R127N and F128Y mutations for cell

recruitment, stably transfected HEK293 cells were tested for their ability to adhere to

CXCL16-expressing ECV304 cells under static conditions. As shown in Figure 43,

changing the DRF motif in CXCR6 did not enable the receptor to mediate adhesion

under static conditions (AI = 0.93; 0.75 and 0.85, respectively).

5 Results

86

!" #$#%& %'()* +'(,-./.

./0

'/.

'/0!"12#34.5#$#6'&12#34.5

728(94

:;<=>?@ABCA;=D

Figure 43: R127N and F128Y do not adhere to CXCL16-ECV304 cells.HEK293 cells expressing the indicated receptor variants were assayed for adhesionunder static conditions to wt-ECV304 or CXCL16-ECV304 cell layer. HEK293cells were stained with CalceinAM for 15 minutes. Excess dye was removed bywashing and cells were suspended to a concentration of 5x105 cells/ml and seededonto wt- or CXCL16-ECV304 cells and after washing, the fluorescence signal ofthe adhering cells was measured. Data are shown as mean ± SD of three indepen-dent experiments.

5 Results

87

6 Discussion

The multi-step process of leukocyte extravasation starts with the rolling of leukocytes

on the activated endothelium, leading to flow-resistant adhesion to transmembrane

adhesion molecules. Subsequently, leukocytes transmigrate through the endothelium

towards a gradient of chemotactic molecules, such as formyl peptides, complement

factors and especially chemokines that belong to the class of cytokines. CX3CL1 and

CXCL16 are exceptional chemokines that are expressed as transmembrane molecules on

the surface of physiologic interfaces such as endothelial or epithelial cells. Both

chemokines can be shed by members of the ADAM family of metalloproteinases,

resulting in a soluble form and are implicated in several pathophysiological situations

such as HIV-infection, renal diseases and atherosclerosis [Tong et al., 2000; Furuichi et

al., 2001; Greaves et al., 2001; Galkina et al., 2007].

Transmembrane CX3CL1 is known to promote flow resistant adhesion of leukocytes

to endothelial or epithelial cells, which is mediated by physical interaction with its

receptor CX3CR1 independently from signaling via Gi-proteins [Fong et al., 1998].

Furthermore, CX3CL1 mediates chemotaxis of CX3CR1-expressing cells, like

monocytes, via Gi-protein activation and CX3CR1 signaling [Imai et al., 1997]. The

present study extends these findings by the detailed analysis of the influence of the

CX3CR1-CX3CL1-interaction during transmigration, internalization, signaling and

cleavage. It could be shown that transmembrane CX3CL1 that is expressed by

endothelial cells is sufficient to promote adhesion and transmigration. Analysis of

receptor mutants indicated that transmigration requires the internalization of the ligand-

receptor-complex as well as signaling of the receptor. Additionally, activity of

ADAM10/17 that shed transmembrane CX3CL1 was needed for efficient transmigration.

Based on the results a model of CX3CL1-mediated leukocyte extravasation is proposed.

This thesis expands previous findings by showing that leukocytes not only employ the

6 Discussion

88

receptor CX3CR1 to bind and adhere to endothelial CX3CL1, but also for transmigration

of adherent leukocytes. This transmigration could be induced by the presence of

transmembrane or soluble CX3CL1, but not further enhanced by the presence of both

forms. Pretreatment of the endothelial cell layer with a neutralizing antibody to CX3CL1

abrogated CX3CL1-induced transmigration of peripheral blood mononuclear cells

(PBMC). Thus, CX3CR1/CX3CL1 interaction was necessary for leukocyte

transmigration. Application of inhibitory antibodies to ICAM-1 and VCAM-1, two major

adhesion molecules involved in leukocyte adhesion, led to a reduced level of CX3CL1-

mediated transmigration. It has been previously reported that inhibition of VCAM-1 on

activated endothelial cells reduced C5a-induced transmigration of monocytes, while

inhibition of ICAM-1 does not [Chuluyan and Issekutz, 1993]. However, CX3CR1

expression on stably transfected L1.2 cells did not activate integrin-mediated adhesion to

VCAM-1 [Haskell et al., 1999], whereas soluble CX3CL1 induced ICAM-1 expression

in endothelial cells [Yang et al., 2007]. The adhesion of CX3CR1 positive cells to

endothelial cells was previously shown to be ICAM-1-dependent [Yang et al., 2007].

Since the pretreatment of endothelial cells with inhibitory antibodies to VCAM-1 and

ICAM-1 abolished CX3CL1-mediated transmigration to a similar extent, not only

ICAM-1 but also VCAM-1 appear to contribute to CX3CL1-induced transmigration.

As in a transmigration assay with stimulated (and thoroughly washed) HUVEC as cell

layer, the transmigration of responsive cells was comparable to that obtained with

ECV304 cells, I asked if sufficient soluble CX3CL1 is produced to mediate

transmigration within the duration of the assay. As was shown by ELISA, the amount of

soluble CX3CL1 after 2 hours in both compartments of the transwell system was

comparable to that produced by unstimulated HUVEC. Since this quantity is not

sufficient to induce chemotaxis of PBMC or L1.2 cells it is unlikely that transmigration

is mediated by soluble CX3CL1. This data indicates a crucial role for transmembrane

CX3CL1 in the transmigration process.

In a transmigration assay with an inhibitory antibody against CX3CL1 that was

6 Discussion

89

applied baolaterally and/or apically, it could be shown that transmigration was

predominantly mediated by apically expressed transmembrane CX3CL1, whereas the

inhibition of basal transmembrane CX3CL1 did not affect transmigration. This result is

consistent with the fact that immobilized chemokines can induce haptotactic migration of

responsive leukocytes, and that a concentration gradient of soluble chemokine is not

required for this activity [Rot and von Andrian, 2004].

Furthermore, the pretreatment of CX3CR1-expressing cells with soluble CX3CL1

abolished CX3CL1-mediated transmigration. This effect was either due to

internalization, and therefore desensitization of the receptor towards CX3CL1, or due to

binding site occupancy.

Endocytosis of ligand-receptor complexes plays a vital role in signal termination and

receptor resensitization, and is mediated by intracellular phosphorylation of the receptor,

and β-arrestin-binding with subsequent uncoupling of G-proteins. Internalization by

clathrin-coated vesicles eventually leads to the dissociation of the ligand and

dephosphorylation of the receptor. It is still under discussion whether internalized

receptor-ligand-pairs still signal or receptor resensitization enhances ligand-induced

signaling. Once the receptor is recycled back to the cell surface, it is resensitized and

competent to signal again [Wolfe and Trejo, 2007]. Dynasore is a small molecule

inhibitor of dynamin, which is essential for the formation of clathrin-coated vesicles. Just

recently it has been shown that dynasore prevents the CCL2-mediated endocytosis of

CCR2 [García Lopez et al., 2009]. Since pretreatment with dynasore inhibited

transmigration towards CX3CL1, it is very likely that internalization of the receptor

plays a crucial role in CX3CR1/CX3CL1-mediated transmigration.

As internalization seemed to be important for transendothelial migration, I next

established receptor variants to analyze the molecular determinants for this receptor

function. Therefore, a highly conserved region in G-protein coupled receptor superfamily

6 Discussion

90

implicated in internalization of the receptor, namely NPX2-3Y in the seventh

transmembrane region, was changed into APLIY and NPLIA. Additionally, a truncation

mutant was constructed that lacked all C-terminal serine residues that can become

phosphorylated and interact with β-arrestins [Barak et al., 1994]. These receptor variants

were then stably transfected into HEK293 and L1.2 cells. Both cell lines did not express

CX3CR1 endogenously. Although HEK293 cells were previously described to migrate

[Dijkstra et al., 2004], no migration of CX3CR1-transfected cells towards CX3CL1

could be observed, but CX3CR1-expressing HEK293 cells adhered as efficiently as

PBMC to CX3CL1-expressing cells. Consistently, CX3CR1-transfected L1.2 cells

showed similar migration properties towards CX3CL1 compared to PBMC. These results

justify the use of both cell lines as a model system for CX3CR1-CX3CL1-interaction.

The mutational analysis of the regions revealed that they are not required for surface

expression, but for ligand binding and trafficking of the receptor, respectively. The

NPX2-3Y motif of CX3CR1 is needed for ligand binding, as it has been reported for

some G-protein coupled receptors, like type 1 angiotensin II receptor [Hunyady et al.,

1995]. In contrast, in other G-protein coupled receptors, like β2-adrenergic receptor, the

NPX2-3Y motif is implicated in internalization, but not ligand binding [Barak et al.,

1995]. Ligand binding takes place at the receptor's N-terminus that ends in the first

transmembrane domain, and also on extracellular loops [Harrison et al., 2001]. Since the

NPX-3Y motif is located in the seventh transmembrane domain, which does not bind the

ligand directly, the loss of ligand binding is most likely due to sterical changes, as the

seventh transmembrane domain is located adjacent to the first [Palczewski et al., 2000].

Most chemoattractant receptors including CXCR2 and CCR5 carry a number of serine

residues that become phosphorylated upon ligand engagement and may trigger the

interaction with β-arrestin which in turn mediates receptor desensitization and

internalization [Ben-Baruch et al., 1995; Kraft et al., 2001]. Accordingly, deletion of the

serine-rich C-terminal part of CX3CR1 attenuated ligand uptake by internalization, but

not calcium signaling. In line with the observation that internalization plays an important

6 Discussion

91

role in the transmigration process, CX3CL1-mediated (trans-)migration of L1.2 cells

expressing the truncation variant of CX3CR1 was impaired. Thus, the C-terminus

contributes to trafficking, but not to induction of intracellular calcium transients.

Lucas and coworkers previously showed that Gi-protein inhibition abrogates

CX3CL1- mediated chemotaxis [Lucas et al., 2003]. Accordingly, pretreatment of

CX3CR1-L1.2 cells with the Gi-protein inhibitor pertussis toxin resulted in the

suppression of transmigration. These findings indicate that CX3CL1-induced

transmigration critically depends on Gi-protein signaling. From previous studies it is

known that the calcium response of G-protein coupled receptors depends on the DRY

motif and its interaction with Gi-proteins [Lagane et al., 2005; Berchiche et al., 2007].

Additionally, the arginine residue of this region has been shown to have effects on

receptor expression, internalization, and binding of the gonadotropin-releasing hormone

receptor [Arora et al., 1997; Ballesteros et al., 1998]. A CX3CR1 variant with a DNY

instead of the DRY motif was constructed, stably expressed in HEK293 and L1.2 cells,

and analyzed. As expected, the induction of intracellular calcium transients upon

treatment with soluble CX3CL1 was abolished in cells expressing the receptor variant.

As (trans-)migration of CX3CR1-R127N expressing cells is impaired, calcium-signaling

is essential for CX3CR1-mediated (trans-)migration. Apparently, this rapid calcium

response occurs independently of slower processes such as receptor internalization.

Taken together, the DRY motif of CX3CR1 is critical for the activation of calcium

signals, but contrary to other reports for different class A GPCR not for receptor

trafficking, as assessed by studying ligand uptake.

Probing of the cell substrate in order to find an appropriate site for transmigration is a

crucial part in the extravasation process of leukocytes and was previously described for

neutrophils in response to the chemoattractant N-formyl-methionyl-leucyl-phenylalanine

[Alteraifi and Zhelev, 1997]. When L1.2 cells were co-incubated with wt- or CX3CL1-

ECV304 cells, L1.2 cells expressing the wildtype receptor extended pseudopods on the

6 Discussion

92

CX3CL1-expressing cell layer. This behavior could not be seen on wt-ECV304 cell

layers or by L1.2 cells expressing no CX3CR1 or the R127N and S319X variants. Both,

intracellular calcium signaling and F-actin network rearrangement are required for

efficient migration, although different signaling pathways seem to be involved. It has

been described that clathrin-deficient cells show an increase in turning and roundness,

and a decrease in polarity, velocity and chemotaxis-efficiency suggesting a role for the

suppression of pseudopod formation and stability [Wessels et al., 2000]. Thus, CX3CR1/

CX3CL1 interaction and the distinct coupled signaling processes are necessary for

pseudopod formation as an initial step of transmigration.

Although the DRY sequence as well as the C-terminus of CX3CR1 differentially

contribute to signaling and trafficking, they are both required for the induction of

chemotaxis and transmigration. The importance of the DRY motif may be explained by

the fact that it represents a critical motif for Gi-protein activation, which triggers

phospholipase C activation followed by diacylglycerol formation and calcium

mobilization. This signaling pathway is involved in the control of actin polymerization,

which is important for cell migration [Murdoch and Finn, 2000; Samstag et al., 2003]. In

murine CX3CR1, the DRY motif may have a similar function, as indicated by the finding

that its mutation into DNY blocks chemotaxis towards soluble CX3CL1 [Haskell et al.,

1999]. The integrity of the receptor’s C-terminus is required for receptor desensitization

via endocytosis, as shown for other chemokine receptors such as CXCR2 [Ludwig et al.,

1997] and clathrin-mediated signaling. This may explain why both Gi-protein activation

by the DRY motif as well as receptor regulation at the C-terminus are critical for

induction of chemotaxis by soluble CX3CL1. The data furthermore demonstrate that

both motifs are required for transmigration in response to transmembrane CX3CL1

suggesting that they may fulfill very similar functions in chemotaxis and transmigration.

Consistent with previous publications showing that adhesion does not require signaling,

adhesion is neither affected by mutating the DRY motif nor by truncation of the C-

6 Discussion

93

terminus [Fong et al., 1998; Haskell et al., 1999].

Transmembrane CX3CL1 can be cleaved proximal to the cell membrane resulting in

the release of its soluble ectodomain and the downregulation of CX3CL1 from the cell

surface [Garton et al., 2001; Hundhausen et al., 2003]. This effect on the one hand leads

to reduced adhesiveness for CX3CR1-expressing leukocytes via transmembrane

CX3CL1. On the other hand, leukocytes bound to transmembrane CX3CL1 may become

released upon cleavage of the transmembrane chemokine [Hundhausen et al., 2007].

Such a process could be relevant for transmigration when adherent leukocytes move

from the apical site of the endothelium towards the lateral junction for transmigration

between two substrate cells and would require the detachment from the apical side

including the interruption of the initial contacts between leukocytes and endothelial cells.

Since the binding of CX3CL1 to its receptor is extremely tight due to a low dissociation

rate from the receptor and a rather high average bond strength, cleavage of CX3CL1

might be the only way to disrupt the cell-cell contact and thereby allow transmigration

[Haskell et al., 2000; Lee et al., 2004].

In the present work, it was demonstrated that activity of ADAM10 and ADAM17 was

required for CX3CL1-induced transmigration. For this purpose, ADAM10 and ADAM17

were inhibited by using pharmacologic inhibitors and shRNA. It could be shown that the

application of the inhibitor led to an increased surface expression and a reduced shedding

of CX3CL1. While the blockade of shedding resulted in an enhanced adhesive activity of

CX3CL1, transmigration of CX3CR1-expressing L1.2 cells towards CX3CL1 was

dramatically reduced, as was shown for ECV304 cells as well as cytokine-stimulated

HUVEC. This data support the model that not only signaling and internalization of the

receptor are crucial for facilitating transmigration, but also the cleaving activity of

ADAM10 and ADAM17. It has been shown earlier that the inhibition of shedding of cell

adhesion molecules, like L-selectin (ADAM17) and VE-Cadherin (ADAM10), decreases

transmigration of T-cells [Faveeuw et al., 2001; Schulz et al., 2008]. Metalloproteinase

activation and activity could be one step in a cascade of events required for lymphocytes

6 Discussion

94

to extravasate the blood vessel. Thereby, ADAMs might not only regulate the

degradation of apical adhesion complexes, but also the loosening of adherens junctions,

as was shown for the VE-cadherin/α-catenin/β-catenin/plakoglobin complex that was lost

from adherens junctions at sites of monocyte penetration of the endothelial layer and

rapidly regained after transmigration [Allport et al., 2000].

Extravasation of leukocytes out of the bloodstream is thought to take place in a multi-

step process that starts with rolling onto endothelial cells via selectins, turns into

chemokine-mediated firm adhesion to adhesion molecules such as integrins, and ends

with the migration of leukocytes towards chemotactic factors through the endothelial

barrier [Springer, 1994]. These events can be mediated by coordinated activity of

selectins, chemokines and integrins. Unlike other chemokines, CX3CL1 has the capacity

to mediate capture under flow, tight adhesion and directional transmigration. Therefore,

it could contribute to all steps of the recruitment process. From this study, I would like to

propose that the process of CX3CL1-mediated cell recruitment involves a sequence of

events starting with physical binding of the receptor, activation of the receptor initiating

the transmigration process, and ending with the cleavage of CX3CL1 by ADAM10 and

ADAM17 allowing the cells to proceed in diapedesis. The last step of transmigration

may involve the cleavage of several surface molecules that contribute to the

transmigration process such as VCAM-1, JAM-A and VE-cadherin, and very likely also

the shedding of transmembrane CX3CL1 [Pruessmeyer and Ludwig, 2008].

6 Discussion

95

Figure 44: Proposed model. Based on the findings presented in this thesis, a mod-el concerning the involvement of CX3CR1/CX3CL1 interaction in the extravasa-tion process can be proposed. Expression of transmembrane CX3CL1 is induced bythe activation of the blood vessel lining endothelium on the luminal as well as theabluminal membrane. While PBMC (and especially monocytes) roll on the en-dothelium due to interactions with selectins, they can be captured not only by clas-sical adhesion molecules, like ICAM-1 and VCAM-1, but also by transmembraneCX3CL1. Acting as a signaling receptor, CX3CR1 might mediate important signalsfor the transmigration process. In order to find an appropriate site to extravasate,PBMC have to probe the tissue they are adhering to. Therefore, bonds have to bedissolved and others closed. Important molecules in dissolving the bond betweenendothelial cell and leukocyte formed by CX3CR1 and CX3CL1 are the metallo-proteases ADAM10 and ADAM17. Additionally, they cleave unoccupied CX3CL1forming a gradient of soluble chemokines and dissolving complexes between en-dothelial cells facilitating leukocyte transmigration.

The data provide a molecular explanation for the proinflammatory activity of the

6 Discussion

96

CX3CR1/CX3CL1 axis in vascular diseases such as atherosclerosis, which is well

documented by a number of recent studies using CX3CR1- or CX3CL1-deficient mice

[Combadière et al., 2003; Lesnik et al., 2003; Teupser et al., 2004]. Finally, when

considering the CX3CR1/CX3CL1 axis as a therapeutic target in order to block

leukocyte recruitment in atherosclerosis, it seems preferable to interfere with the initial

step of adhesion using a receptor antagonist rather than blocking CX3CR1 signaling or

CX3CL1 shedding, which would both lead to the accumulation of adherent leukocytes at

the vascular surface.

In the future a set of further experiments should be conducted to elucidate the role of

CX3CR1/CX3CL1 in the extravasation process. It is known that β-arrestins are

important for the internalization of G-protein coupled receptors and subsequent signaling

and interact with the intracellular part of the receptor [Kraft et al., 2001]. Additionally,

some reports show that CX3CL1 and CX3CR1 are involved in survival and proliferation

signals in microglia, epithelial cells and smooth muscle cells in a NF-κb-dependent

manner [Boehme et al., 2000; Brand et al., 2002; Chandrasekar et al., 2003]. Since both,

the DRY sequence as well as the C-terminus of CX3CR1 are required for transmigration,

the analysis of other pathways would possibly lead to a better understanding of signaling

pathways, other than intracellular calcium flux, involved in CX3CR1/CX3CL1 function.

There are two possibilities to analyze the involvement of CX3CR1/CX3CL1-mediated

adhesion and transmigration: first, the construction of a signaling-deficient receptor

variant and second the construction of a non-cleavable ligand variant. As it is known that

ADAMs are able to employ a vast number of different cleavage sites, the construction of

a non-cleavable ligand variant seemed to be more difficult than that of receptor variants.

Still, it would be a great help to have a CX3CL1 variant that can not be cleaved, and this

project should be pursued with different approaches.

As was shown earlier, the CX3CR1/CX3CL1 axis is involved in the progression of

6 Discussion

97

inflammatory diseases like atherosclerosis [McDermott et al., 2001; Moatti et al., 2001]

and nephropathies [Furuichi et al., 2001; Cockwell et al., 2002]. In further experiments

the CX3CR1 variants should be analyzed in inflammatory models in vivo. Therefore, the

receptor has to be reconstituted in CX3CR1-deficient mice by either knock-in technique

or by lentiviral transfer. This will help to evaluate the magnitude and manner of

contribution of CX3CR1/CX3CL1 in the extravasation process of leukocytes during

inflammation.

CXCL16 and its receptor CXCR6 are implicated in a multitude of pathophysiological

situations as atherosclerosis and HIV infection [Deng et al., 1997; Matloubian et al.,

2000; Galkina et al., 2007]. Recent data from studies using mice deficient in either

CXCR6 or CXCL16 support a role for both ligand and receptor in atherogenesis

[Aslanian and Charo, 2006; Galkina et al., 2007].

When peripheral blood mononuclear cells (PBMC) of healthy donors were tested for

expression of CXCR6, it was surprising that the expression of CXCR6 strongly varied

depending on the donor and cell type. Consequently, PBMC from CXCR6low-

expressing donors did not migrate in response to CXCL16 whereas PBMC from donors

showing an expression of CXCR6 migrated towards CXCL16. Surprisingly, unlike

published earlier [Shimaoka et al., 2004], CXCR6-expressing cells did not adhere to

CXCL16-expressing cells. Just recently it was shown that stimulated macrophages do not

adhere to coronary artery smooth muscle cells in a CXCL16-dependent manner under

static conditions [Barlic et al., 2009]. The differential receptor expression and the non-

adherence of CXCR6-positive cells raised the question whether the adhesion of CXCR6-

expressing cells to CXCL16-expressing cells is relevant under physiological conditions.

For other adhesion molecules, shear stress on endothelial cells is important for the

mediation of firm adhesion. Shear stress has been shown to modulate the adhesive

capacity of endothelial cells by adjusting the expression of adhesion molecules as for

example ICAM-1 [McKinney et al., 2006]. Therefore, flow assays are needed to

6 Discussion

98

elucidate the possibility that CXCL16 only induces adhesion under flow conditions.

Furthermore, a network of interactions might be necessary for the CXCR6-CXCL16

interaction to become relevant under physiological conditions.

Instead of the highly conserved DRY motif found in many chemokine receptors,

CXCR6 bears a DRF motif within the second intracellular loop. This DRF motif is also

found in the 5-HT2βPan -receptor of Panulirus interruptus (spiny lobster) and implicated in

its constitutive activity [Clark et al., 2004]. The DRF motif in the Gonadotropin releasing

hormone receptor (GnRH-R) of Callithrix jacchus (common marmoset) mediates a

slightly faster internalization rate than the DRS motif found in GnRH-R of most

mammals [Byrne et al., 1999]. When the DRF motif of CXCR6 was changed into the

conventional DRY, only slight changes in calcium signaling could be observed, whereas

the mutation into DNF abolished the calcium signal. As seen for CX3CR1, the arginine

residue seems to play the crucial role in the mediation of calcium signals, whereas the

tyrosine residue does not contribute. The tyrosine residue is believed to be a potential

phosphorylation site, acting as a substrate for a G-protein receptor kinase, and thus is

important for receptor desensitization and internalization [Palczewski, 1997]. This

suggestion was supported by the findings of a study carried out by Arora and colleagues,

which demonstrated that mutating the serine to a tyrosine in the mouse GnRH-R

increased the rate of receptor internalization and agonist binding affinity, although no

effect was observed on G-protein coupling [Arora et al., 1995]. However, it has also been

shown that generating a mouse GnRH-R DRS to DRA mutation had no significant

effects on ligand binding, receptor coupling or internalization [Arora et al., 1997].

Therefore, for functional analysis the next step should be the investigation of

internalization rates after binding of CXCL16 to CXCR6, similar to the experiments

done for CX3CR1 variants.

6 Discussion

99

Additionally, knock-down of CXCR6 led to the formation of medullablastoma,

suggesting a role in cancer development [Sasai et al., 2007]. The interaction of CXCR6

and CXCL16 is also thought to mediate cell recruitment in rheumatoid arthritis, and

CXCR6 acts as secondary co-receptor for all HIV-strains [Liao et al., 1997; Ruth et al.,

2006]. In upcoming experiments the interaction of CXCR6 and CXCL16 should be

further characterized. Therefore, L1.2 cell expressing CXCR6 and its variants, consistent

with the experiments for CX3CR1 in this thesis, should be generated and analyzed in

(trans)-migration and other functional assays, like phosphorylation of signaling

molecules or mediation of proliferation/survival.

While CXCR6 is thought to be a pro-atherogenic chemokine receptor, as its knock-

down resulted in a decrease of susceptibility to atherosclerosis, CXCL16 might be

atheroprotective due to the fact that its knock-down resulted in accelerated

atherosclerosis, which has been associated with the additional scavenger receptor

function of the chemokine. CXCL16 not only binds to CXCR6 to mediate cell

recruitment, but also acts as scavenger receptor for oxidized low density lipoproteins.

This function of CXCL16 is not influenced by CXCR6. The uptake of low density

lipoproteins eventually leads to an upregulation of atheroprotective genes, such as ATP-

binding cassette transporter-1 and apolipoprotein E [Liao et al., 2002; Barlic et al., 2009].

This supports the hypothesis that CXCL16 mediates atheroprotection not through the

interaction of CXCR6 and CXCL16, but through its scavenger function. Antagonism of

the CXCR6-CXCL16 interaction without affecting the scavenger receptor activity may

be of therapeutic potential.

Although both chemokines exist as soluble as well as transmembrane forms, the

functions they mediate on binding to their receptors seem to differ. While CX3CR1

mediates adhesion as well as migration, CXCR6 only mediates migration. The

6 Discussion

100

experiments showed that transmembrane CX3CL1 contributes to the extravasation of

leukocytes, while there is no contribution of transmembrane CXCL16. To obtain a better

understanding of the underlying mechanisms receptor variants were generated. Although

the CX3CR1 (and to a lesser degree the CXCR6) variants were extensively

characterized, some questions remain open and should be addressed in a future set of

experiments. The importance of the C-terminal part of CX3CR1 was shown in the

context of receptor desensitization but it is still unclear what pathways and functions

exactly are affected. Therefore, experiments targeting the receptor phosphorylation at

intracellular serine-residues, receptor recycling or F-actin network rearrangement should

be performed. Additionally, the unusual DRF motif in CXCR6 might influence the

receptor's activity turning it into a constitutively active receptor in contrast to receptors

bearing the conserved DRY motif. The consequence of such a potential constitutive

activity in the context of chemotaxis should be investigated. Furthermore, the impact of

CX3CL1 and CXCL16 shedding still remains unclear even though it was shown that the

activity of the sheddases is required for efficient CX3CL1-mediated transmigration. With

the variety of mediated functions, the CX3CR1-CX3CL1 axis, and possibly also

CXCR6-CXCL16, may constitute important regulators involved in leukocyte

extravasation.

As all the presented data were generated in vitro, these results should be confirmed by

in vivo experiments. Despite their different structural characteristics and activity,

CX3CR1 and CXCR6 show a pro-atherogenic potential as confirmed by usage of knock-

out mice for CX3CR1 and CXCR6 [Combadière et al., 2003; Galkina et al., 2007]. These

mice can now be reconstituted with the mutated murine receptors by knock-in technique

or lentiviral genetransfer as is currently established in our laboratory. The impact of the

alteration of conserved motifs in CX3CR1 and CXCR6 should be investigated in the

context of inflammatory animal models, such as the ApoE-/- diet-induced atherosclerosis

or wire-induced injury model. In the latter model, CX3CL1 is upregulated in intimal

smooth muscle cells and endothelial cells. Furthermore, CX3CR1 deficiency is

6 Discussion

101

associated with a decreased infiltration of monocytes [Liu et al., 2006]. The relevance of

CX3CR1-CX3CL1 axis in humans is underlined by the fact that the CX3CR1

polymorphism V249I is associated with increased monocyte adhesiveness increasing the

risk for restenosis after coronary stent implantation [Daoudi et al., 2004; Niessner et al.,

2005]. Accordingly, the alteration of the DRY motif of CX3CR1 or truncation of the C-

terminal serine-residues can have the following impact in a murine model of

inflammation: cells expressing the CX3CR1 variants, that both block transmigration,

should accumulate on the arterial wall and the infiltration into the lesion should be

reduced.

Similarly, in a model of ApoE-/- diet-induced atherosclerosis, mice deficient in CXCR6

display reduced atherosclerosis associated with a lower content of CXCR6+ T cells and

macrophages in the aorta [Galkina et al., 2007]. Since CXCR6-dependent chemotaxis

requires signaling, less cells expressing the receptor variants should be recruited into the

lesion, when the DRF motif is abolished. In contrast to CX3CL1, CXCL16 does not

mediate adhesion. Therefore, as indicated by the in vitro experiments, cells expressing

the mutated DRF motif of CXCR6 should not accumulate at the vascular wall. Thus,

mutating either receptor should result in a less severe inflammatory response. This may

be complicated by the fact that mutated CX3CR1 still functions as adhesion molecule

leading to an accumulation of CX3CR1-variant expressing cells at the arterial wall. A

pharmacological interference with transmembrane CX3CL1 should block both, the

accumulation of leukocytes at the site of inflammation and their infiltration into the

lesion.

6 Discussion

102

7 Literature

1. Abel, S. et al. The transmembrane CXC-chemokine ligand 16 is induced by IFN-gamma and TNF-alpha and shed by the activity of the disintegrin-likemetalloproteinase ADAM10. J Immunol 172, 6362-6372 (2004).

2. Acharya, S. & Karnik, S. S. Modulation of GDP release from transducin by theconserved Glu134-Arg135 sequence in rhodopsin. J Biol Chem 271, 25406-25411(1996).

3. Alewijnse, A. E. et al. The effect of mutations in the DRY motif on the constitutiveactivity and structural instability of the histamine H(2) receptor. Mol Pharmacol57, 890-898 (2000).

4. Allen, S. J., Crown, S. E. & Handel, T. M. Chemokine: receptor structure,interactions, and antagonism. Annu Rev Immunol 25, 787-820 (2007).

5. Allport, J. R., Muller, W. A. & Luscinskas, F. W. Monocytes induce reversiblefocal changes in vascular endothelial cadherin complex during transendothelialmigration under flow. J Cell Biol 148, 203-216 (2000).

6. Alteraifi, A. M. & Zhelev, D. V. Transient increase of free cytosolic calcium duringneutrophil motility responses. J Cell Sci 110, 1967-1977 (1997).

7. Arora, K. K., Cheng, Z. & Catt, K. J. Mutations of the conserved DRS motif in thesecond intracellular loop of the gonadotropin-releasing hormone receptor affectexpression, activation, and internalization. Mol Endocrinol 11, 1203-1212 (1997).

8. Arora, K. K., Sakai, A. & Catt, K. J. Effects of second intracellular loop mutationson signal transduction and internalization of the gonadotropin-releasing hormonereceptor. J Biol Chem 270, 22820-22826 (1995).

9. Aslanian, A. M. & Charo, I. F. Targeted disruption of the scavenger receptor andchemokine CXCL16 accelerates atherosclerosis. Circulation 114, 583-590 (2006).

10. Ballesteros, J. et al. Functional microdomains in G-protein-coupled receptors. Theconserved arginine-cage motif in the gonadotropin-releasing hormone receptor. JBiol Chem 273, 10445-10453 (1998).

7 Literature

103

11. Barak, L. S., Ménard, L., Ferguson, S. S., Colapietro, A. M. & Caron, M. G. Theconserved seven-transmembrane sequence NP(X)2,3Y of the G-protein-coupledreceptor superfamily regulates multiple properties of the beta 2-adrenergic receptor.Biochemistry 34, 15407-15414 (1995).

12. Barak, L. S., Oakley, R. H., Laporte, S. A. & Caron, M. G. Constitutive arrestin-mediated desensitization of a human vasopressin receptor mutant associated withnephrogenic diabetes insipidus. Proc Natl Acad Sci USA 98, 93-98 (2001).

13. Barak, L. S. et al. A highly conserved tyrosine residue in G protein-coupledreceptors is required for agonist-mediated beta 2-adrenergic receptor sequestration.J Biol Chem 269, 2790-2795 (1994).

14. Barlic, J., Zhu, W. & Murphy, P. M. Atherogenic lipids induce high-densitylipoprotein uptake and cholesterol efflux in human macrophages by up-regulatingtransmembrane chemokine CXCL16 without engaging CXCL16-dependent celladhesion. J Immunol 182, 7928-7936 (2009).

15. Barreiro, O., Vicente-Manzanares, M., Urzainqui, A., Yanez-Mo, M. & Sanchez-Madrid, F. Interactive protrusive structures during leukocyte adhesion andtransendothelial migration. Front Biosci 9, 1849-1863 (2004).

16. Bazan, J. F. et al. A new class of membrane-bound chemokine with a CX3C motif.Nature 385, 640-644 (1997).

17. Beglova, N., Blacklow, S. C., Takagi, J. & Springer, T. A. Cysteine-rich modulestructure reveals a fulcrum for integrin rearrangement upon activation. Nat StructBiol 9, 282-287 (2002).

18. Ben-Baruch, A. et al. Interleukin-8 receptor beta. The role of the carboxyl terminusin signal transduction. J Biol Chem 270, 9121-9128 (1995).

19. Berchiche, Y. A. et al. Direct assessment of CXCR4 mutant conformations revealscomplex link between receptor structure and G(alpha)(i) activation. J Biol Chem282, 5111-5115 (2007).

20. Boehme, S. A., Lio, F. M., Maciejewski-Lenoir, D., Bacon, K. B. & Conlon, P. J.The chemokine fractalkine inhibits Fas-mediated cell death of brain microglia. JImmunol 165, 397-403 (2000).

21. Bokoch, G. M. Chemoattractant signaling and leukocyte activation. Blood 86,1649-1660 (1995).

7 Literature

104

22. Boulday, G., Coupel, S., Coulon, F., Soulillou, J. P. & Charreau, B. Antigraftantibody-mediated expression of metalloproteinases on endothelial cells.Differential expression of TIMP-1 and ADAM-10 depends on antibody specificityand isotype. Circ Res 88, 430-437 (2001).

23. Brand, S., Sakaguchi, T., Gu, X., Colgan, S. P. & Reinecker, H. C. Fractalkine-mediated signals regulate cell-survival and immune-modulatory responses inintestinal epithelial cells. Gastroenterology 122, 166-177 (2002).

24. Byrne, B. et al. Isolation and characterisation of the marmoset gonadotrophinreleasing hormone receptor: Ser(140) of the DRS motif is substituted by Phe. JEndocrinol 163, 447-456 (1999).

25. Campbell, J. J. et al. Chemokines and the arrest of lymphocytes rolling under flowconditions. Science 279, 381-384 (1998).

26. Chandrasekar, B., Bysani, S. & Mummidi, S. CXCL16 signals via Gi,phosphatidylinositol 3-kinase, Akt, I kappa B kinase, and nuclear factor-kappa Band induces cell-cell adhesion and aortic smooth muscle cell proliferation. J BiolChem 279, 3188-3196 (2004).

27. Chandrasekar, B. et al. Fractalkine (CX3CL1) stimulated by nuclear factor kappaB(NF-kappaB)-dependent inflammatory signals induces aortic smooth muscle cellproliferation through an autocrine pathway. Biochem J 373, 547-558 (2003).

28. Chen, Y., Green, S. R., Almazan, F. & Quehenberger, O. The amino terminus andthe third extracellular loop of CX3CR1 contain determinants critical for distinctreceptor functions. Mol Pharmacol 69, 857-865 (2006).

29. Chuluyan, H. E. & Issekutz, A. C. VLA-4 integrin can mediate CD11/CD18-independent transendothelial migration of human monocytes. J Clin Invest 92,2768-2777 (1993).

30. Clark, M. C. et al. Arthropod 5-HT2 receptors: a neurohormonal receptor indecapod crustaceans that displays agonist independent activity resulting from anevolutionary alteration to the DRY motif. J Neurosci 24, 3421-3435 (2004).

31. Cockwell, P., Chakravorty, S. J., Girdlestone, J. & Savage, C. O. Fractalkineexpression in human renal inflammation. J Pathol 196, 85-90 (2002).

7 Literature

105

32. Combadiere, C., Gao, J., Tiffany, H. L. & Murphy, P. M. Gene cloning, RNAdistribution, and functional expression of mCX3CR1, a mouse chemotacticreceptor for the CX3C chemokine fractalkine. Biochem Biophys Res Commun 253,728-732 (1998).

33. Combadière, C. et al. Decreased atherosclerotic lesion formation in CX3CR1/apolipoprotein E double knockout mice. Circulation 107, 1009-1016 (2003).

34. Daoudi, M. et al. Enhanced adhesive capacities of the naturally occurring Ile249-Met280 variant of the chemokine receptor CX3CR1. J Biol Chem 279,19649-19657 (2004).

35. del Pozo, M. A., Sanchez-Mateos, P., Nieto, M. & Sanchez-Madrid, F. Chemokinesregulate cellular polarization and adhesion receptor redistribution duringlymphocyte interaction with endothelium and extracellular matrix. Involvement ofcAMP signaling pathway. J Cell Biol 131, 495-508 (1995).

36. Deng, H. K., Unutmaz, D., KewalRamani, V. N. & Littman, D. R. Expressioncloning of new receptors used by simian and human immunodeficiency viruses.Nature 388, 296-300 (1997).

37. Dichmann, S. et al. Fractalkine induces chemotaxis and actin polymerization inhuman dendritic cells. Inflamm Res 50, 529-533 (2001).

38. Dijkstra, I. M., Hulshof, S., van der Valk, P., Boddeke, H. W. & Biber, K. Cuttingedge: activity of human adult microglia in response to CC chemokine ligand 21. JImmunol 172, 2744-2747 (2004).

39. Fanelli, F., Barbier, P., Zanchetta, D., de Benedetti, P. G. & Chini, B. Activationmechanism of human oxytocin receptor: a combined study of experimental andcomputer-simulated mutagenesis. Mol Pharmacol 56, 214-225 (1999).

40. Faure, S. et al. Rapid progression to AIDS in HIV+ individuals with a structuralvariant of the chemokine receptor CX3CR1. Science 287, 2274-2277 (2000).

41. Faveeuw, C., Preece, G. & Ager, A. Transendothelial migration of lymphocytesacross high endothelial venules into lymph nodes is affected by metalloproteinases.Blood 98, 688-695 (2001).

42. Feng, L. et al. Prevention of crescentic glomerulonephritis byimmunoneutralization of the fractalkine receptor CX3CR1 rapid communication.Kidney Int 56, 612-620 (1999).

7 Literature

106

43. Fong, A. M. et al. Fractalkine and CX3CR1 mediate a novel mechanism ofleukocyte capture, firm adhesion, and activation under physiologic flow. J Exp Med188, 1413-1419 (1998).

44. Foussat, A. et al. Deregulation of the expression of the fractalkine/fractalkinereceptor complex in HIV-1-infected patients. Blood 98, 1678-1686 (2001).

45. Furuichi, K. et al. Upregulation of fractalkine in human crescenticglomerulonephritis. Nephron 87, 314-320 (2001).

46. Galkina, E. et al. CXCR6 Promotes Atherosclerosis by Supporting T-Cell Homing,Interferon-{gamma} Production, and Macrophage Accumulation in the AorticWall. Circulation (2007).

47. García Lopez, M. A., Aguado Martínez, A., Lamaze, C., Martínez-A, C. & Fischer,T. Inhibition of dynamin prevents CCL2-mediated endocytosis of CCR2 andactivation of ERK1/2. Cell Signal (2009).

48. Garcia, G. E. et al. NF-kappaB-dependent fractalkine induction in rat aorticendothelial cells stimulated by IL-1beta, TNF-alpha, and LPS. J Leukoc Biol 67,577-584 (2000).

49. Garin, A., Pellet, P., Deterre, P., Debré, P. & Combadière, C. Cloning andfunctional characterization of the human fractalkine receptor promoter regions.Biochem J 368, 753-760 (2002).

50. Garton, K. J. et al. Tumor necrosis factor-alpha-converting enzyme (ADAM17)mediates the cleavage and shedding of fractalkine (CX3CL1). J Biol Chem 276,37993-38001 (2001).

51. Geiger, B. & Bershadsky, A. Exploring the neighborhood: adhesion-coupled cellmechanosensors. Cell 110, 139-142 (2002).

52. Gosling, J. et al. Molecular uncoupling of C-C chemokine receptor 5-inducedchemotaxis and signal transduction from HIV-1 coreceptor activity. Proc Natl AcadSci USA 94, 5061-5066 (1997).

53. Greaves, D. R. et al. Linked chromosome 16q13 chemokines, macrophage-derivedchemokine, fractalkine, and thymus- and activation-regulated chemokine, areexpressed in human atherosclerotic lesions. Arterioscler Thromb Vasc Biol 21,923-929 (2001).

7 Literature

107

54. Harrison, J. K. et al. Mutational analysis of the fractalkine chemokine domain.Basic amino acid residues differentially contribute to CX3CR1 binding, signaling,and cell adhesion. J Biol Chem 276, 21632-21641 (2001).

55. Haskell, C. A., Cleary, M. D. & Charo, I. F. Molecular uncoupling of fractalkine-mediated cell adhesion and signal transduction. Rapid flow arrest of CX3CR1-expressing cells is independent of G-protein activation. J Biol Chem 274,10053-10058 (1999).

56. Haskell, C. A., Cleary, M. D. & Charo, I. F. Unique role of the chemokine domainof fractalkine in cell capture. Kinetics of receptor dissociation correlate with celladhesion. J Biol Chem 275, 34183-34189 (2000).

57. Hofnagel, O., Luechtenborg, B., Plenz, G. & Robenek, H. Expression of the novelscavenger receptor SR-PSOX in cultured aortic smooth muscle cells and umbilicalendothelial cells. Arterioscler Thromb Vasc Biol 22, 710-711 (2002).

58. Hundhausen, C. et al. The disintegrin-like metalloproteinase ADAM10 is involvedin constitutive cleavage of CX3CL1 (fractalkine) and regulates CX3CL1-mediatedcell-cell adhesion. Blood 102, 1186-1195 (2003).

59. Hundhausen, C. et al. Regulated shedding of transmembrane chemokines by thedisintegrin and metalloproteinase 10 facilitates detachment of adherent leukocytes.J Immunol 178, 8064-8072 (2007).

60. Hunyady, L., Bor, M., Baukal, A. J., Balla, T. & Catt, K. J. A conserved NPLFYsequence contributes to agonist binding and signal transduction but is not aninternalization signal for the type 1 angiotensin II receptor. J Biol Chem 270,16602-16609 (1995).

61. Imai, T. et al. Identification and molecular characterization of fractalkine receptorCX3CR1, which mediates both leukocyte migration and adhesion. Cell 91, 521-530(1997).

62. Kao, J. P., Harootunian, A. T. & Tsien, R. Y. Photochemically generated cytosoliccalcium pulses and their detection by fluo-3. J Biol Chem 264, 8179-8184 (1989).

63. Kim, C. H. et al. Bonzo/CXCR6 expression defines type 1-polarized T-cell subsetswith extralymphoid tissue homing potential. J Clin Invest 107, 595-601 (2001).

64. Kim, J. H., Cho, E. Y., Min, C., Park, J. H. & Kim, K. M. Characterization offunctional roles of DRY motif in the 2nd intracellular loop of dopamine D2 and D3receptors. Arch Pharm Res 31, 474-481 (2008).

7 Literature

108

65. Kim, K. M. & Caron, M. G. Complementary roles of the DRY motif and C-terminus tail of GPCRS for G protein coupling and beta-arrestin interaction.Biochem Biophys Res Commun 366, 42-47 (2008).

66. Kraft, K. et al. Characterization of sequence determinants within the carboxyl-terminal domain of chemokine receptor CCR5 that regulate signaling and receptorinternalization. J Biol Chem 276, 34408-34418 (2001).

67. Krueger, K. M., Daaka, Y., Pitcher, J. A. & Lefkowitz, R. J. The role ofsequestration in G protein-coupled receptor resensitization. Regulation of beta2-adrenergic receptor dephosphorylation by vesicular acidification. J Biol Chem 272,5-8 (1997).

68. Lagane, B. et al. Mutation of the DRY motif reveals different structuralrequirements for the CC chemokine receptor 5-mediated signaling and receptorendocytosis. Mol Pharmacol 67, 1966-1976 (2005).

69. Landsman, L. et al. CX3CR1 is required for monocyte homeostasis andatherogenesis by promoting cell survival. Blood 113, 963-972 (2009).

70. Laporte, S. A. et al. The beta2-adrenergic receptor/betaarrestin complex recruits theclathrin adaptor AP-2 during endocytosis. Proc Natl Acad Sci U S A 96, 3712-3717(1999).

71. Lee, F. H., Haskell, C., Charo, I. F. & Boettiger, D. Receptor-ligand binding in thecell-substrate contact zone: a quantitative analysis using CX3CR1 and CXCR1chemokine receptors. Biochemistry 43, 7179-7186 (2004).

72. Lee, S. J. et al. Fractalkine stimulates angiogenesis by activating the Raf-1/MEK/ERK- and PI3K/Akt/eNOS-dependent signal pathways. Am J Physiol Heart CircPhysiol 291, H2836-46 (2006).

73. Lesnik, P., Haskell, C. A. & Charo, I. F. Decreased atherosclerosis in CX3CR1-/-mice reveals a role for fractalkine in atherogenesis. J Clin Invest 111, 333-340(2003).

74. Liao, F. et al. STRL33, A novel chemokine receptor-like protein, functions as afusion cofactor for both macrophage-tropic and T cell line-tropic HIV-1. J Exp Med185, 2015-2023 (1997).

75. Liao, H., Langmann, T., Schmitz, G. & Zhu, Y. Native LDL upregulation of ATP-binding cassette transporter-1 in human vascular endothelial cells. ArteriosclerThromb Vasc Biol 22, 127-132 (2002).

7 Literature

109

76. Liu, P. et al. CX3CR1 deficiency confers protection from intimal hyperplasia afterarterial injury. Arterioscler Thromb Vasc Biol 26, 2056-2062 (2006).

77. Lucas, A. D. et al. Smooth muscle cells in human atherosclerotic plaques expressthe fractalkine receptor CX3CR1 and undergo chemotaxis to the CX3C chemokinefractalkine (CX3CL1). Circulation 108, 2498-2504 (2003).

78. Ludwig, A. et al. Enhanced expression and shedding of the transmembranechemokine CXCL16 by reactive astrocytes and glioma cells. J Neurochem 93,1293-1303 (2005).

79. Ludwig, A., Berkhout, T., Moores, K. E., Groot, P. & Chapman, G. Fractalkine isexpressed by smooth muscle cells in response to IFN-gamma and TNF-alpha and ismodulated by metalloproteinase activity. J Immunol 168, 604-612 (2002).

80. Ludwig, A. et al. Metalloproteinase inhibitors for the disintegrin-likemetalloproteinases ADAM10 and ADAM17 that differentially block constitutiveand phorbol ester-inducible shedding of cell surface molecules. Comb Chem HighThroughput Screen 8, 161-171 (2005).

81. Ludwig, A. et al. The CXC-chemokine neutrophil-activating peptide-2 induces twodistinct optima of neutrophil chemotaxis by differential interaction withinterleukin-8 receptors CXCR-1 and CXCR-2. Blood 90, 4588-4597 (1997).

82. Macia, E. et al. Dynasore, a cell-permeable inhibitor of dynamin. Dev Cell 10,839-850 (2006).

83. Man, S., Ubogu, E. E. & Ransohoff, R. M. Inflammatory cell migration into thecentral nervous system: a few new twists on an old tale. Brain Pathol 17, 243-250(2007).

84. Martin-Padura, I. et al. Junctional adhesion molecule, a novel member of theimmunoglobulin superfamily that distributes at intercellular junctions andmodulates monocyte transmigration. J Cell Biol 142, 117-127 (1998).

85. Matloubian, M., David, A., Engel, S., Ryan, J. E. & Cyster, J. G. A transmembraneCXC chemokine is a ligand for HIV-coreceptor Bonzo. Nat Immunol 1, 298-304(2000).

86. McDermott, D. H. et al. Chemokine receptor mutant CX3CR1-M280 has impairedadhesive function and correlates with protection from cardiovascular disease inhumans. J Clin Invest 111, 1241-1250 (2003).

7 Literature

110

87. McDermott, D. H. et al. Association between polymorphism in the chemokinereceptor CX3CR1 and coronary vascular endothelial dysfunction andatherosclerosis. Circ Res 89, 401-407 (2001).

88. McKinney, V. Z., Rinker, K. D. & Truskey, G. A. Normal and shear stressesinfluence the spatial distribution of intracellular adhesion molecule-1 expression inhuman umbilical vein endothelial cells exposed to sudden expansion flow. JBiomech 39, 806-817 (2006).

89. Minta, A., Kao, J. P. & Tsien, R. Y. Fluorescent indicators for cytosolic calciumbased on rhodamine and fluorescein chromophores. J Biol Chem 264, 8171-8178(1989).

90. Mizoue, L. S., Bazan, J. F., Johnson, E. C. & Handel, T. M. Solution structure anddynamics of the CX3C chemokine domain of fractalkine and its interaction with anN-terminal fragment of CX3CR1. Biochemistry 38, 1402-1414 (1999).

91. Moatti, D. et al. Polymorphism in the fractalkine receptor CX3CR1 as a geneticrisk factor for coronary artery disease. Blood 97, 1925-1928 (2001).

92. Moore, C. A., Milano, S. K. & Benovic, J. L. Regulation of receptor trafficking byGRKs and arrestins. Annu Rev Physiol 69, 451-482 (2007).

93. Muller, W. A., Weigl, S. A., Deng, X. & Phillips, D. M. PECAM-1 is required fortransendothelial migration of leukocytes. J Exp Med 178, 449-460 (1993).

94. Murdoch, C. & Finn, A. Chemokine receptors and their role in inflammation andinfectious diseases. Blood 95, 3032-3043 (2000).

95. Nakayama, T. et al. Cutting edge: profile of chemokine receptor expression onhuman plasma cells accounts for their efficient recruitment to target tissues. JImmunol 170, 1136-1140 (2003).

96. Niessner, A. et al. Fractalkine receptor polymorphisms V2491 and T280M asgenetic risk factors for restenosis. Thromb Haemost 94, 1251-1256 (2005).

97. Nieto, M. et al. Polarization of chemokine receptors to the leading edge duringlymphocyte chemotaxis. J Exp Med 186, 153-158 (1997).

98. Nishida, N. et al. Activation of leukocyte beta2 integrins by conversion from bentto extended conformations. Immunity 25, 583-594 (2006).

99. Onuffer, J. J. & Horuk, R. Chemokines, chemokine receptors and small-moleculeantagonists: recent developments. Trends Pharmacol Sci 23, 459-467 (2002).

7 Literature

111

100. Palczewski, K. et al. Crystal structure of rhodopsin: A G protein-coupled receptor.Science 289, 739-745 (2000).

101. Palczewski, K. GTP-binding-protein-coupled receptor kinases--two mechanisticmodels. Eur J Biochem 248, 261-269 (1997).

102. Passam, A. M. et al. Polymorphisms of Cx(3)CR1 and CXCR6 Receptors inRelation to HAART Therapy of HIV Type 1 Patients. AIDS Res Hum Retroviruses23, 1026-1032 (2007).

103. Pruessmeyer, J. & Ludwig, A. The good, the bad and the ugly substrates forADAM10 and ADAM17 in brain pathology, inflammation and cancer. Semin CellDev Biol (2008).

104. Rosenthal, W., Antaramian, A., Gilbert, S. & Birnbaumer, M. Nephrogenicdiabetes insipidus. A V2 vasopressin receptor unable to stimulate adenylyl cyclase.J Biol Chem 268, 13030-13033 (1993).

105. Ross, R. The pathogenesis of atherosclerosis: a perspective for the 1990s. Nature362, 801-809 (1993).

106. Rostene, W., Kitabgi, P. & Parsadaniantz, S. M. Chemokines: a new class ofneuromodulator? Nat Rev Neurosci 8, 895-903 (2007).

107. Rot, A. et al. RANTES and macrophage inflammatory protein 1 alpha induce themigration and activation of normal human eosinophil granulocytes. J Exp Med 176,1489-1495 (1992).

108. Rot, A. & von Andrian, U. H. Chemokines in innate and adaptive host defense:basic chemokinese grammar for immune cells. Annu Rev Immunol 22, 891-928(2004).

109. Ruth, J. H. et al. CXCL16-mediated cell recruitment to rheumatoid arthritissynovial tissue and murine lymph nodes is dependent upon the MAPK pathway.Arthritis Rheum 54, 765-778 (2006).

110. Samstag, Y., Eibert, S. M., Klemke, M. & Wabnitz, G. H. Actin cytoskeletaldynamics in T lymphocyte activation and migration. J Leukoc Biol 73, 30-48(2003).

111. Sasai, K. et al. Medulloblastomas derived from Cxcr6 mutant mice respond totreatment with a smoothened inhibitor. Cancer Res 67, 3871-3877 (2007).

112. Sato, T. et al. Role for CXCR6 in recruitment of activated CD8+ lymphocytes toinflamed liver. J Immunol 174, 277-283 (2005).

7 Literature

112

113. Scheer, A. et al. Mutational analysis of the highly conserved arginine within theGlu/Asp-Arg-Tyr motif of the alpha(1b)-adrenergic receptor: effects on receptorisomerization and activation. Mol Pharmacol 57, 219-231 (2000).

114. Schenkel, A. R., Mamdouh, Z., Chen, X., Liebman, R. M. & Muller, W. A. CD99plays a major role in the migration of monocytes through endothelial junctions. NatImmunol 3, 143-150 (2002).

115. Schulz, B. et al. ADAM10 regulates endothelial permeability and T-Celltransmigration by proteolysis of vascular endothelial cadherin. Circ Res 102,1192-1201 (2008).

116. Sharron, M. et al. Expression and coreceptor activity of STRL33/Bonzo on primaryperipheral blood lymphocytes. Blood 96, 41-49 (2000).

117. Shimaoka, T. et al. Molecular cloning of a novel scavenger receptor for oxidizedlow density lipoprotein, SR-PSOX, on macrophages. J Biol Chem 275,40663-40666 (2000).

118. Shimaoka, T. et al. Cell surface-anchored SR-PSOX/CXC chemokine ligand 16mediates firm adhesion of CXC chemokine receptor 6-expressing cells. J LeukocBiol 75, 267-274 (2004).

119. Slice, L. W. et al. The conserved NPXnY motif present in the gastrin-releasingpeptide receptor is not a general sequestration sequence. J Biol Chem 269,21755-21761 (1994).

120. Springer, T. A. Traffic signals for lymphocyte recirculation and leukocyteemigration: the multistep paradigm. Cell 76, 301-314 (1994).

121. Stocker, W. & Bode, W. Structural features of a superfamily of zinc-endopeptidases: the metzincins. Curr Opin Struct Biol 5, 383-390 (1995).

122. Tanaka, Y., Adams, D. H. & Shaw, S. Proteoglycans on endothelial cells presentadhesion-inducing cytokines to leukocytes. Immunol Today 14, 111-115 (1993).

123. Teupser, D. et al. Major reduction of atherosclerosis in fractalkine (CX3CL1)-deficient mice is at the brachiocephalic artery, not the aortic root. Proc Natl AcadSci USA 101, 17795-17800 (2004).

124. Tong, N. et al. Neuronal fractalkine expression in HIV-1 encephalitis: roles formacrophage recruitment and neuroprotection in the central nervous system. JImmunol 164, 1333-1339 (2000).

7 Literature

113

125. Wessels, D. et al. Clathrin plays a novel role in the regulation of cell polarity,pseudopod formation, uropod stability and motility in Dictyostelium. J Cell Sci113, 21-36 (2000).

126. Wilbanks, A. et al. Expression cloning of the STRL33/BONZO/TYMSTRligandreveals elements of CC, CXC, and CX3C chemokines. J Immunol 166, 5145-5154(2001).

127. Wilbanks, A. M., Laporte, S. A., Bohn, L. M., Barak, L. S. & Caron, M. G.Apparent loss-of-function mutant GPCRs revealed as constitutively desensitizedreceptors. Biochemistry 41, 11981-11989 (2002).

128. Wolfe, B. L. & Trejo, J. Clathrin-dependent mechanisms of G protein-coupledreceptor endocytosis. Traffic 8, 462-470 (2007).

129. Wolfsberg, T. G. & White, J. M. ADAMs in fertilization and development. DevBiol 180, 389-401 (1996).

130. Wong, S. K. G protein selectivity is regulated by multiple intracellular regions ofGPCRs. Neurosignals 12, 1-12 (2003).

131. Yang, X. P. et al. Fractalkine Upregulates Intercellular Adhesion Molecule-1 inEndothelial Cells Through CX3CR1 and the Jak Stat5 Pathway. Circ Res (2007).

132. Zhu, S. Z., Wang, S. Z., Hu, J. & el-Fakahany, E. E. An arginine residue conservedin most G protein-coupled receptors is essential for the function of the m1muscarinic receptor. Mol Pharmacol 45, 517-523 (1994).

7 Literature

114

8 List of figures

Figure 1: Multiple steps are necessary for leukocyte extravasation. 8................................

Figure 2: Chemokines are divided into four groups. 11........................................................

Figure 3: Schematic view of CX3CR1. 17............................................................................

Figure 4: Schematic view of CX3CR1 signaling. 19............................................................

Figure 5: Structure of GW280264X. 30................................................................................

Figure 6: Structure of Dynasore. 31......................................................................................

Figure 7: CX3CR1 is expressed on PBMC. 48.....................................................................

Figure 8: PBMC adhere to a CX3CL1 expressing cell layer under flow conditions. 49......

Figure 9: PBMC chemotactically migrate towards CX3CL1. 50..........................................

Figure 10: PBMC chemotactically transmigrate in response to CX3CL1. 51......................

Figure 11: Stimulated HUVEC express CX3CL1. 52...........................................................

Figure 12: Neutralizing of CX3CL1, ICAM-1 and VCAM-1 prevents transmigration. 53..

Figure 13: Dynasore prevents transmigration of PBMC. 54.................................................

Figure 14: Cells migrate towards CX3CL1 and adhere to CX3CL1. 56...............................

Figure 15: L1.2 cells transmigrate through an ECV304 cell layer. 56..................................

Figure 16: L1.2 cells transmigrate through a HUVEC cell layer. 57....................................

Figure 17: HUVEC produce CX3CL1 after stimulation with cytokines. 58........................

Figure 18: CX3CL1 inhibition influences transmigration. 59...............................................

Figure 19: PTX treatment influences transmigration. 60......................................................

Figure 20: Pretreatment with dynasore prevents L1.2 cell chemotaxis. 61...........................

Figure 21: Alignment of conserved motifs in GPCR superfamily. 62..................................

Figure 22: Overview of introduced changes in CX3CR1. 63...............................................

Figure 23: Expression and binding of CX3CR1 variants. 64................................................

Figure 24: Stably transfected HEK293 cells. 65...................................................................

Figure 25: CX3CR1 internalizes and mediates CX3CL1 uptake. 66....................................

Figure 26: CX3CR1 and S319X, but not R127N mediate intracellular calcium flux. 68.....

Figure 27: Receptor variants mediate adhesion under static and flow conditions. 69..........

8 List of figures

115

Figure 28: Receptor variants do not mediate (trans-)migration. 71......................................

Figure 29: Receptor variant R127N does not mediate transmigration through HUVEC.72

Figure 30: Receptor variants do not mediate pseudopod formation. 73................................

Figure 31: Inhibition of ADAM10/17 abrogates transmigration. 75....................................

Figure 32: CX3CL1 surface expression is reduced when incubated with PBMC. 75..........

Figure 33: PBMC adhere, but do not transmigrate when ADAM10/17 are inhibited. 76....

Figure 34: PBMC adhere, but do not transmigrate when ADAM10/17 are inhibited. 77....

Figure 35: ADAM10 silencing leads to suppression of transmigration. 78........................

Figure 36: CXCR6 expression varies based on donor. 81.....................................................

Figure 37: PBMC do not adhere to CXCL16 under static conditions. 81.............................

Figure 38: CXCR6-expressing cells do not adhere to CXCL16 under static conditions. 82

Figure 39: CXCR6 binds CXCL16-6xHis. 83......................................................................

Figure 40: PBMC expressing CXCR6 chemotactically migrate towards CXCL16. 84........

Figure 41: Surface expression and ligand binding of CXCR6 variants. 85..........................

Figure 42: CXCR6 and F128Y, but not R127N mediate intracellular calcium flux. 86.......

Figure 43: R127N and F128Y do not adhere to CXCL16-ECV304 cells. 87.......................

Figure 44: Proposed model. 96..............................................................................................

8 List of figures

116

9 List of tablesTable 1: Conserved motifs in GPCR class A. 15...................................................................

9 List of tables

117

10 Abbreviations

AADAM a disintegrin and metalloproteinaseADP adenosine diphosphateAF AlexaFluorAI adhesion indexAIDS acquired immunodeficiency syndromeAM acetoxymethylesterAmp ampicillinANOVA analysis of varianceapoE apolipoprotein EATP adenosine-5'-triphosphateBBad Bcl-2-associated death promoter Bcl B-cell leukemiaBSA bovine serum albumineCCAM cellular adhesion moleculecAMP cyclic adenosine monophosphateCD chemokine domaincDNA complementary DNACTAK cutaneous T-cell-attracting chemokineDDAG diacylglycerolDNA desoxyribonucleic acidDMEM Dulbecco's modified Eagle's MediumDMSO dimethyl sulfoxidedNTP desoxy nucleoside triphosphateEE.coli Escherichia coliEDTA N,N'-1,2-ethanediylbis(N-(carboxymethyl)glycine)edetic acidEGTA (Ethylenebis(oxyethylenenitrilo))tetra-; ethylene glycol bis(2-

aminoethyl ether)-N,N,N'N'tetraacetic acidE-selectin endothelial selectin

10 Abbreviations

118

ELISA enzyme-linked immunosorbent assayeNOS endothelial nitric oxide synthaseERK extracellular-signal regulated kinaseFFACS fluorescence-activated cell sorterFCS fetal calf serumFig. figureGg standard gravityG418 geneticin sulfateGDP guanosine diphosphateGi-protein inhibitory G-proteinGnRH-R gonadotropin-releasing hormone receptorGTP guanosine-5'-triphosphateGPCR G-protein coupled receptorHh hourHAART highly active anti-retroviral therapyHBSS Hank's buffered salt solutionHEPES 4-2-hydroxyethyl-1-piperazineethanesulfonic acid HCl hydrochloric acidHIV human immunodeficiency virusHRP horseradish peroxidase5-HT2βPan 5-hydroxytryptamin receptor 2β PanHUVEC human umbilical vein endothelial cellsIICAM intercellular adhesion moleculeIC50 half maximal inhibitory concentrationIFN interferonIgG immunglobulin GI-κB inhibitors of NF-κBI-κK I-κB kinase complexIL interleukinIP3 inositol 1,4,5-trisphosphate

10 Abbreviations

119

JJAM junctional adhesion moleculeJNK c-Jun N-terminal kinasesKkb kilo base pairskDa kilo DaltonLLFA lymphocyte function associated antigenLPS lipopolysaccharideL-selectin leukocyte selectinMM molarmAb monoclonal antibodyMac-1 macrophage antigen-1MAP kinase mitogen-activated protein kinaseMI migration indexmin minuteMIP macrophage inflammatory proteinmRNA messenger ribonucleic acidNNaCl sodium chlorideNF-κB nuclear factor 'kappa-light-chain-enhancer' of activated B-cellsOORF open reading frameox-LDL oxidized low density lipoproteinPPBMC peripheral blood mononuclear cellPBS phosphate buffered salinePCR polymerase chain reactionPE phycoerythrinPECAM platelet/endothelial cell adhesion moleculePen/Strep penicillin/streptomycinPFA paraformaldehydepfu Pyrococcus furiosusPI3K phosphoinositide 3-kinase

10 Abbreviations

120

PIP2 phosphatidylinositol 4,5-bisphosphatePKB proteinkinase BPKC proteinkinase CPLC phospholipase CPMA phorbol-12-myristat-13-acetatPOD peroxidaseP-selectin platelet selectinPTX pertussis toxinRRANTES regulated upon activation, normal T-cell expressed, and secretedRNA ribonucleic acidrpm rounds per minuteRPMI roswell park memorial instituteRT room temperatureSs secondSAPK stress-activated phospho-kinasesSD standard deviationSDS sodiumdodecylsulfateSH3 Src-homology 3shRNA short interfering RNASIV simian immunodeficiency virusesSrc sarcomaSR-PSOX scavenger receptor that binds phosphatidylserine and oxidized

lipoproteinSOB super optimal brothTTACE tumor necrosis factor alpha converting enzymeTaq Thermus aquaticusTECK thymus-expressed chemokineTI transmigration indexTNF tumor necrosis factorUU unit

10 Abbreviations

121

VVCAM vascular cell adhesion moleculeVE-Cadherin vascular endothelial cadherinVLA-4 very late antigen 4Wwt wildtypeXX-gal bromo-chloro-indolyl-galactopyranoside

A alanine (Ala) G glycine (Gly) M methionine (Met) S serine (Ser)C cysteine (Cys) H histidine (His) N asparagine (Asn) T threonine (Thr)D aspartic acid (Asp) I isoleucine (Ile) P proline (Pro) V valine (Val)E glutamic acid (Glu) K lysine (Lys) Q glutamine (Gln) W tryptophane

(Trp)F phenylalanine (Phe) L leucine (Leu) R arginine (Arg) Y tyrosine (Tyr)

10 Abbreviations

122

11 Vectors

11.1 hCX3CR1 in pcDNA3.1+

11 Vectors

123

11.2 hCX3CR1 R127N in pcDNA3.1+

11 Vectors

124

11.3 hCX3CR1 N289A in pcDNA3.1+

11 Vectors

125

11.4 hCX3CR1 Y293A in pcDNA3.1+

11 Vectors

126

11.5 hCX3CR1 S319X in pcDNA3.1+

11 Vectors

127

11.6 hCXCR6 in pcDNA3.1+

11 Vectors

128

11.7 hCXCR6 R127N in pcDNA3.1+

11 Vectors

129

11.8 hCXCR6 F128Y in pcDNA3.1+

11 Vectors

130

12 Curriculum Vitae

Personal data

Birth date 29.10.1981

Birth place Strausberg, Germany

Nationality german

School education

1994 - 2001 Angelagymnasium Osnabrück

05/2001 Allgemeine Hochschulreife

University education

10/2001 - 07/2004 Bachelor: Molecular Biotechnology, University of Lübeck

04/2004 - 07/2004 Bachelor-Thesis: Etablierung des Modells der beschichtetenKapillare für in vitro Untersuchungen von mononukleären Zellenunter Flussbedingungen

08/2004 Bachelor of Science in Molecular Biotechnology

10/2004 - 06/2006 Master: Biomedical Engineering, Hochschule Anhalt and Martin-Luther-University Halle-Wittenberg

10/2005 - 06/2006 Master-Thesis: Einfluss von Lymphoid Enhancer Factor (Lef-1)auf den Zellzyklus von Kardiomyozyten

06/2006 Master of Engineering in Biomedical Engineering

from 07/2006 PhD-student, RWTH Aachen

from 07/2006 PhD-Thesis: Requirements for leukocyte transendothelialmigration via the transmembrane chemokines CX3CL1 andCXCL16

12 Curriculum Vitae

131

13 Publications

Articles

N. Schwarz, J. Pruessmeyer, F. M. Hess, E. Pantaler, R. Windoffer, M. Voss, A. Sarabi, C.Weber, A. Sechi, S. Uhlig, A. LudwigRequirements for leukocyte transendothelial migration via the transmembranechemokine CX3CL1submitted

J. Pruessmeyer, C. Martin, F. M. Hess, N. Schwarz, S. Schmidt, T. Kogel, N. Hoettecke,B. Schmidt, A. Sechi, S. Uhlig, A. LudwigThe disintegrin and metalloproteinase 17 (ADAM17) mediates inflammation-induced shedding of syndecan-1 and -4 by lung epithelial cells J Biol Chem. 2009 Oct 29. [Epub ahead of print]

R. R. Koenen, J. Pruessmeyer, O. Soehnlein, L. Fraemohs, A. Zernecke, N. Schwarz, K. Reiss, A. Sarabi, L. Lindbom, T. M. Hackeng, C. Weber, A. LudwigRegulated release and functional modulation of junctional adhesion molecule A by disintegrin metalloproteinasesBlood. 2009 May 7;113(19):4799-809

C. Hundhausen, A. Schulte, B. Schulz, M.G. Andrzejewski, N.Schwarz, P. vonHundelshausen, U. Winter, K. Paliga, K. Reiss, P. Saftig, C. Weber, A. LudwigRegulated Shedding of Transmembrane Chemokines by the Disintegrin andMetalloproteinase 10 Facilitates Detachment of Adherent LeukocytesJ Immunol. 2007 Jun 15;178(12):8064-72

13 Publications

132

Poster

Regulated Cleavage of transmembrane chemokines by ADAM10 and ADAM17.M. G. Andrzejewski, N. Schwarz, C. Weber, A. Ludwig37th Symposium of the Germany Society for Immunology (DGfI), Heidelberg, 2007Symposium of the Bonner Forum for Biomedicine, Bonn, 2007

Distinct structural determinants are required for CX3CR1-mediated cell adhesionand chemotaxisN. Schwarz, E. Pantaler, A. LudwigSymposium of the Austrian Society for Allergology/Immunology, together withGermany Society for Immunology (DGfI), Wien, Österreich, 2008Society for Microcirculation and Vascular Biology, Aachen, 2008

Sequential steps of leukocyte recruitment via the transmembrane chemokineCX3CL1N. Schwarz, J. Pruessmeyer, F.M. Hess, R. Windoffer, M. Voss, A. Sarabi, C. Weber, A.Sechi, S. Uhlig, A. Ludwig2nd European Congress of Immunology, Berlin, 2009Signal Transduction and Disease, Aachen, 2009

The proinflammatory cytokines IFNγ and TNFα induce shedding of endothelial andepithelial surface molecules via upregulation of ADAM17 activityJ. Pruessmeyer, N. Schwarz, F.M. Hess, T. Kogel, C. Martin, S. Uhlig, A. Ludwig2nd European Congress of Immunology, Berlin, 2009Signal Transduction and Disease, Aachen, 2009

13 Publications

133

14 Declaration

I hereby declare, that this thesis was carried out at RWTH Aachen University, within

the Institute for Pharmacology and Toxicology and the Institute for Molecular and

Cardiovascular Research. It was exclusively performed by myself, unless otherwise

stated in the text. To my knowledge, it contains no material used in other publications or

thesis, except where reference is made in the text.

_______________________________________ Aachen, 14. Januar 2010

14 Declaration

134

15 Acknowledgements

Ich möchte mich bei all denjenigen Personen bedanken, die zum Gelingen dieser

Dissertation beigetragen und mich unterstützt haben.

An erster Stelle geht mein Dank an PD Dr. Andreas Ludwig. Vielen Dank für die

Geduld, die exzellente Betreuung und die Begutachtung der Doktorarbeit. Darüber

hinaus, danke für bitter benötigten Kaffee und Tee in allen Lebenslagen, danke für jede

Menge offener Ohren bei absurden Gedanken, danke für die investierte Zeit, die ab und

an sicher viele Nerven gekostet hat.

Ich danke Prof. Stefan Uhlig für die unkomplizierte Aufnahme in seine Arbeitsgruppe.

Ausserdem vielen Dank für die Übernahme des Gutachtens.

Prof. Werner Baumgartner möchte ich für die Übernahme des Gutachtens ohne langes

Nachdenken danken. Mir ist ein riesiger Stein vom Herzen gefallen.

Mein Dank geht an Jessica Prüßmeyer für die vielen fruchtbaren Gespräche bei

Käffchen oder Schnitzel sowohl auf privater als auch professioneller Ebene.

Besonderer Dank geht an unsere Chef-TA Tanja Kogel und ihre Vize Melanie Esser.

Danke für die Geduld, den Spaß und vor allem die Unterstützung. Let's do the dynasore

rock!

Danke an Daniela Dreymüller für das Erschlagen der kleinen Fehlerteufelchen.

Vergiss unser Date im Sommer nicht. ;)

Ganz besonders möchte ich mich bei meiner Familie bedanken. Ihr wart während all

der Jahre immer eine große Unterstützung und habt nie aufgehört, an mich zu glauben.

Und zum Schluss noch ein ganz großes Danke an Patrick Beier für die Hilfe mit den

Bildern, das Nutzen teurer Satelliten um Tupperdosen im Wald zu suchen und vorallem

die Zeit, die wir miteinander verbracht haben und in Zukunft verbringen werden.

Danke!

15 Acknowledgements

135

Requirements for leukocyte transendothelial migration via the

Documents

Transcript of Requirements for leukocyte transendothelial migration via the